Inventory of Roman London: Structures outside the walls

An Inventory of the Historical Monuments in London, Volume 3, Roman London. Originally published by His Majesty's Stationery Office, London, 1928.

This free content was digitised by double rekeying. All rights reserved.


'Inventory of Roman London: Structures outside the walls', in An Inventory of the Historical Monuments in London, Volume 3, Roman London( London, 1928), British History Online [accessed 19 July 2024].

'Inventory of Roman London: Structures outside the walls', in An Inventory of the Historical Monuments in London, Volume 3, Roman London( London, 1928), British History Online, accessed July 19, 2024,

"Inventory of Roman London: Structures outside the walls". An Inventory of the Historical Monuments in London, Volume 3, Roman London. (London, 1928), , British History Online. Web. 19 July 2024.

(C) Structures Outside the Walls.

(a) North of the River.

Aldersgate Street. A wall was found in 1887 on the N. side of St. Botolph's churchyard, the substructure of which was thought to be Roman [Arch., XXX, 522; Soc. Ant. MS. Min. XXXVIII, 206].

Bishopsgate Street (Without). A City Sewers Plan of 1852 (III, 12) has the following note of discoveries in Montague Court:—"While excavating for a new sewer 30 ft. from the frontage of Bishopsgate Street we met with an old wall 9 ft. from the surface to the top of the wall; it was 3 ft. thick, built with Kentish rag and mortar with two courses of flat tiles every foot in height; it ran in a parallel line with Bishopsgate Street. Twenty feet farther up the court we met with another wall parallel with the last and built in the same manner, 8 ft. to the top of the same and 3 ft. thick."

Fig. 58. Pile-structure, Blomfield Street. From Archæological Journal LX, by permission.

Blomfield Street. In 1901, considerable remains of pile-structures (Plate 53 and Figs. 58–61, were discovered on the W. side of the street and on the W. side of the bed of the Walbrook. The structures consisted of piles and cross-beams and planking; some of the beams were grooved and tenoned. The space between the boarding was apparently rubbish brought from elsewhere to form a platform and perhaps added to during the occupation. The evidence from all these structures was consistent and showed that 18 in. of sand and silt had been deposited on the original gravel surface of the river-bed before their erection. The objects found on the site afforded direct evidence of Roman date, none of the pottery figured being assignable to a date later than A.D. 130. The pottery was found in, and mostly at the top of, the filling forming the platforms. The discoveries were fully reported on by F. W. Reader [Arch. Journ., LX, 137–204 and 313–335].

Fig. 59. Pile structures, Blomfield Street. From Arch. Journ., LX.

In 1925, when the site of Nos. 13 and 14 on the E. side of Blomfield Street was excavated, Mr. Q. Waddington observed a number of piles, perhaps forming part of the eastern embankment of the stream. Some lst-century pottery was also found.

Holborn. A fragment of geometric pavement in black, red, and white was found in the 17th century near St. Andrew's Church. It was taken up, given to the Royal Society, and long preserved in the Museum at Gresham College [N. Grew, Museum Regalis Societatis (1681), 380].

Hyde Park. The original Ossulston Stone, said to have been a Roman "geometric stone," formerly stood near the N.E. corner of the Park on the S. side of Oxford Street; it is marked on Rocque's map as "milestone." The stone was subsequently dug up and placed against the Marble Arch, but has now disappeared [Lond. and Midd. Arch. Soc. Trans., IV, 62].

Long Lane (Smithfield). Mr. T. Fisher records the discovery, before 1805, of a pavement in this locality, but no further particulars are given [Gent. Mag., 1807, I, 415–7].

Old Bond Street. In March, 1894, a stone culvert with joints of brick, set in cement, was found, running southwards [Antiq., XXIX, 244].

St. Martin in the Fields. During the excavations for the building of this church, 1722, "a Roman brick arch was found with several ducts, 14 ft. underground, and Sir Hans Sloane had a bell-shaped glass vase that was found in a stone coffin among ashes in digging the foundations of the portico." Stukeley described it as "an arch built of Roman brick and at the bottom laid with a most strong cement. . . . There was a square duct in each wall its whole length, of 9 in. breadth; there were several of these side by side." [Gough, Camden, II, 17, 93; Brayley, Beauties of Engl, and Wales, X, Pt. I, 91; Allen. Hist. of London, I, 25; Soc. Antiq. MS. Min. I, 151, 170; Arch. Rev., I, 356].

Strand Lane, E. side. About 80 yards S. of the Strand, is a plunge-bath of brick with a round N. end, 15½ ft. by 6¾ ft. This bath was formerly lined with modern marble slabs, and has only recently been stripped; the actual walls and floor, where they can be tested, are built of red bricks, 9 in. by 4¾ in. by 1¾ in. The bath is fed by a spring which now enters at the S.E. corner, and there are remains of the former feed-pipe or overflow in the middle of the S. end. The date of the bath is at present uncertain; the bricks are unlike any Roman bricks yet discovered in this country, but, on the other hand, they do not resemble in form or texture the normal bricks of the 17th century, which seems to be the alternative date.

Fig. 60. Pile-structure, Blomfield Street. From Arch. Journ. LX.

Fig. 61. Pile-structures, Blomfield Street. Sections of coarse pottery-rims found. From Arch. Journ. LX.

Water Street. A note on a City Sewers Plan in 1849 reads: "In this spot (the N. part of Water Street extending into Tudor Street) we met with a row of oak piles 3 ft. apart, with oak planking nailed on the front of them, 12 in. wide and 1½ in. thick. The piles were 12 in. square and 10 ft. from the surface to the top of do—50 ft. in length." There was apparently no evidence of the date of this structure.

Westminster Abbey Precincts. In digging the foundations of new Canons' houses in the Abbey garden in 1883, remains of a Roman "dwelling" were found at a depth of 14 ft.; they consisted of slabs of concrete flooring, roof-tiles and other rubbish. Similar remains to those last described are said to have been discovered also in the cloister [Arch. Journ.. XLII. 274]. In 1878, when digging the grave of Sir Gilbert Scott, Roman building material was found under the nave of the Abbey church, said to have been remains of the pilae of a hypocaust. A Roman sarcophagus of Oxfordshire oolite was found in 1869 on the N. side of the Abbey church; it is now in the chapter-house vestibule. It measures 6 ft. 10 in. by 2 ft. 4 in. by 1 ft. 10 in. (see p. 173).

(b) South of the River. (See Plan B. p. 150).

Castle Street, Southwark. Brock's map marks "hypocaust flues marked Px Tx" between this street and Barclay and Perkins' Brewery. The find is probably identical with one recorded by Taylor as on the latter site [Annals of St. Mary Overy, 10; see Park Street].

Deverell Street, Kent Road. A Roman "hypocaust or flue" was found about 1825 near the Dissenters' burial-ground [Gent. Mag., 1825, II, 633].

Guildford Street, Southwark. Numerous piles were found here about 1867 [Journ. Brit. Arch. Assoc., XXIII, 87].

High Street, Southwark. In making the Southwark approach to New London Bridge about 1830, a Roman pavement of coarse tesserae was found in the middle of Borough High Street [Arch., XXIV, 198].

In 1840, on the W. side about 100 yards N. of St. George's Church (Plan B 1) were found flue and roof-tiles, Gaulish and other pottery, beads, fragments of glass bottles, a bell, coins of Tiberius, Faustina I, Severus and Tetricus, and frescopaintings of a superior kind. Some of the lastnamed had foliage and flowers in green, yellow, and white on a dark ground, others plain borders of red, green and white [Arch., XXIX, 149].

At King's Head Yard (Plan B 3), in 1879–80, were found a fragment of tessellated pavement, a coin of Domitian, and pottery [Journ. Brit. Arch. Assoc., XXXVI, 122, 234]. Further excavations on the same site in 1881 yielded more results: flue and roof-tiles, fragments of stamped amphoræ, Gaulish pottery and other varieties, a key, and coins of Vespasian and Domitian. A coin of Justinus (A.D. 537) was also reported. These remains appear to betoken the presence of an inhabited building [Ibid., XXXVII, 211, 427].

Fig. 62. Pavement under Borough High Street. From Arch., LXIII.

In 1908, in cutting for a drain between No. 52 and the sewer (Plan B 2) a pavement (Fig. 62) of roughly shaped rag-stones was found at a depth of 21 ft. from the surface. The pavement was supported on piles driven into the sand, and above it were found fragments of Roman pottery, some of which was of late 1st and early 2nd-century date [Arch., LXIII, 323, with site-plan and section; Trans. Croyd. N. H. and Sc. Soc., 1911, 44].

Mint Street, Southwark. Finds here included, in 1887, two clay water-pipes [Journ. Brit. Arch. Assoc., XLIII, 374].

Mitre Street, Southwark. A clay water-pipe from this street is in the Guildhall Museum [Cat., 120].

Park Street, Southwark (Plan B 11). In 1658, Dugdale records the discovery of "a Roman pavement made of bricks not above an inch and a half square, and adjoining to it a more curious piece of the like small bricks, in length about 10 ft. and in breadth 5 ft., wrought in various colours; and in the midst thereof, betwixt certain borders in the fashion of wreathed columns, the form of a serpent very lively expressed in that kind of Mosaic work" [Dugdale, History of Imbanking and Drayning, 65; hence Gwilt in Gent. Mag., 1815, I, 225]. In 1806, Taylor records the find of a flue-tile inscribed Px Tx [Annals of St. Mary Overy, p. 10, pl. 1, fig. 3]. Timber, nails and iron-work were found in December, 1868, at the corner of Clink street (Plan B10) from "an ancient wooden structure formed of stout piles set about two feet apart and supporting beams and joists overlaid with planking rabbeted and fastened again by broad-headed four-sided nails of iron," supposed to be a Roman landing-place [Journ. Brit. Assoc., XXV, 79].

St. George's Fields, Southwark. "In these fields, commonly called St. George's, many Roman coins, tessellated work, bricks and rubble have been found from time to time, also a large urn, full of bones" [T. Gale, Antonini Iter, 1709, 65]. "Tessellated pavements and urns" are said to have been found in St. George's Fields [Gent. Mag., 1825, I, 148].

St. Saviour, Southwark. Brock's map marks on the S. side of the church (Plan B 8) a mosaic pavement found 18 July, 1820; also "a footpath of red Roman." Lindsay says it was in the court in front of St. Saviour's School [Etym. of Southwark, 3rd Ed., 5]. In 1825, Gwilt found fragments of a pavement and a quantity of Roman bricks worked into the walls [Gent. Mag., 1825, II, 633; Lindsay, Etym. of Southwark, 6; Taylor, Annals of St. Mary Overy, 15]. In 1831 stone foundations were found under the choir of the church and running N.E. and S.W. [Arch. XXIV, 198]. In 1833, part of a tessellated pavement was found in the churchyard [Gent. Mag., 1833, I, 255].

In 1839, in digging for foundations of warehouses round the church, traces of walls were found together with tesserae, frescoes, etc. Partly on the site of St. Saviour's Grammar School (S. of the church, Plan B 9), and partly under the adjoining house a tessellated pavement of a handsome pattern was found and in the churchyard nearly opposite was a narrow pavement of red tesserae running from N.E. to S.W. [Arch., XXIX, 148; Gent. Mag., 1840, I, 192].

About 1910 a portion of red tessellated pavement was found at the S.E. corner of the churchyard (Plan B 7) when digging a trench for a new railing. The pavement was 9½ ft. below the surface and rested on 16 in. of builders' rubbish. It probably formed part of the same pavement as that discovered in 1833 [Arch., LXIII, 325, with section].

St. Thomas Watering (Old Kent Road). A Janus-head in marble (figured in Allen's History of London, I, 36) was dug up about 1690, near this point, together with large flat bricks and other Roman remains [Woodward's Letter to Hearne]. According to Defoe it was found in connection with remains of a building and a second head was also found and left in quicksand [D. Defoe, Tour through Britain, I, 234].

St. Thomas's Hospital, Southwark, site of. In 1840, on pulling down the S. wing of the outer or western quadrangle of St. Thomas's Hospital (Plan B 6) "a Roman pavement of the common red tesserae, surrounded by walls (Plate 46) of flint and rubble with courses of Roman tiles has been discovered at a depth of 20 ft. from the level of the High Street. The pavement measured about 20 ft. by 12 ft.; the tesserae were embedded in concrete about 6 in. thick under which was a layer of chips of stone. On removing the foundations of the walls they were found to rest on piles, the soil being sand .... we were informed that on the N. side there were the jambs of a doorway and on the W. side a continuation of the buildings." Roach Smith records that on the floor there were found several coins of the Constantine family [Gent. Mag., 1840, I, 191–2; Arch., XXIX, 148 with plan, etc., pl. 18].

St. Thomas's Street, Southwark. A tessellated pavement was found at the corner of High Street (Plan B 5), in 1819, at a depth of 10 ft. [Brock's map].

Southwark Street. A tessellated pavement was discovered in 1820 on the site of Cure's College (Plan B 12) i.e. between Park Street and Southwark Street [Surrey Arch. Colls., XXVIII, 141].

In excavating for the formation of Southwark Street (Plan B 13) in 1862 numerous remains came to light, including fragments of tessellated pavements and wall-paintings at a depth of from 10 to 26 ft. [Lond. and Midd. Arch. Soc. Trans., II, Proceedings, 84]. In 1866, on the S. side of the street between Southwark Square and Worcester Street (Plan B 14), "in a space of about 100 ft. by 40 ft., 16 pits had been sunk, each disclosing Roman pottery above a number of piles and puddled clay." The piles were 7 to 11 ft. in length, the heads about 12 ft. below the street-level. On the opposite side of the street (Plan B 15), 135 ft. from the piling, a pavement of red tesserae and remains of Roman pottery were found [Journ. Brit. Arch. Assoc., XXII, 445 ff.; XXIII, 87; Lond. and Midd. Arch. Soc. Trans., III, 213, n.].

Stoney Street, Southwark. Where this street now joins High Street (Plan B 4), Brock's map marks the discovery of red stucco.

(c) Structures in outlying parts of the County of London.

Blackheath. Part of a rectangular earthwork is marked on the O.S. 25 inch London cv. not far S. of Greenwich Park, and on Blackheath near Hollyhedge House. Excavations were made in it in 1906 when Roman tiles and coarse pottery were found in the mound and ditch and within the enclosure but not outside its boundaries. [Information given by Mr. Herbert Jones to Prof. Haverfield, and now preserved among the Vict. Co. Hist. MS. materials for Kent to which access was kindly given by Mr. William Page, the Editor.] Professor Haverfield suggested that the site may have been a temple enclosure or villa garden, but that in its present shape it is too imperfect to be taken into account as a determinant feature. No doubt it is to be connected with the structure found in Greenwich Park.

Charlton. The earthwork at Charlton was situated on the edge of the river-marsh about ½ mile N.E. of Charlton church. It was a fort of the contour class and occupied the end of a low spur rising to rather above the 100 ft. contour. Its form, not now ascertainable with certainty, was probably an irregular pentagon with rounded angles, covering with its defences an area of perhaps 17½ acres. The greater part of the earthwork was destroyed in the 18th and early 19th century in digging for sand, but a considerable stretch of the W. side together with the southern angle survived until 1870, when it was surveyed by Sir Flinders Petrie [Arch. Cant., XIII]. Between that date and 1915 the greater part of the surviving banks and ditches had been destroyed. In the latter year excavations were undertaken by Mr. Elliston Erwood to determine the date and character of the earthwork. The defences on the W. side consisted of two banks and inner and medial ditches; the inner ditch was formed by scarping the edge of the plateau. The small surviving portion of the enclosed area contained a number of habitationfloors indicating the former existence of hutments commonly of roughly circular form. The finds on the site consisted of flint-flakes, furnace-bars, querns, loom-weights, fibulae, pottery and two coins of Claudius. All the pottery and fibulae appeared to date between A.D. 60 and 250, with the exception of one fragment of a bronze-age urn. The only masonry structure was found by workmen in 1906, and described by them as a round building of about 20 sq. ft. area, and with walls of flint, etc., standing 2½ ft. high.

The settlement would appear to have been one inhabited by Romano-British natives established within an earthwork of uncertain date, but probably not greatly, if at all, anterior to the settlement [Journ. Brit. Arch. Assoc., N.S., XXII, 125].

A small subsequent excavation in 1923, and the continued destruction of the remains of the earthwork provided evidence extending the date of the occupation to the end of the 3rd and possibly into the 4th century. Mr. Erwood considers that this settlement is the Noviomagus of the Antonine Iter II [Journ. Brit. Arch. Assoc., N.S., XXIX, 227.]

Greenwich. In 1902, the remains of a Roman building were partially excavated in the N. of Greenwich Park, about 100 yds. from its E. wall and half way between Vanbrugh and Maze Hill gates, at the point where the high ground of the main area of the Park reaches its northern limit. The plan of the building (which may or may not have been a shrine) was not ascertained; only a small piece of rag-stone walling and three patches of flooring—two of opus signinum and one, 3 ft. higher than the others, of coarse red tesserae— were found, scattered over an area about 15 by 50 ft. Much burnt matter, tiles, hypocaust bricks, painted wall-plaster, cubes from a mosaic, a piece of green porphyry, worked and moulded blocks of oolite, parts of drums of three small columns, and some window-glass were also unearthed. The small finds were numerous, amongst them about 300 Roman coins ranging from Claudius to Honorius and including perhaps 200 of "Constantinian copper." A legionary denarius of Mark Antony (Leg. XIV) was dug up about 100 yds. N.E. of the main site.

The most important relics, however, were:—

(i) Part of a figure (Plate 51) in oolite about two-thirds life-size, showing a right arm with elbow bent and turned back to the body; on the forearm is an armlet, and drapery falls over the shoulder to the hand.

(ii) Sandstone fragment, 6½ in. by 8 in., bearing the edges of the two first lines of an inscription in letters 1¾ in. tall.

C V L A P or R


In the first line Aesculapius has been conjectured. [Eph. Ep., IX, 992]. In the British Museum.

(iii) White marble fragment, 6 in. high by 4½ in. wide, with letters 7/8 in. high, from the left side of an inscription.




S I or V

Possibly Numinibus Augusti, but conjecture is useless [Eph. Ep., IX, 993]. In the British Museum.

(iv) White marble, bearing the tops of three letters which might be A S S.

(v–vii) Three other fragments bearing parts of letters [H. Jones, Home Counties Mag., V, 49, 213 (plan); A. D. Webster, Greenwich Park (Greenwich, 1902), 67–100, figs., not agreeing exactly in detail; Daily Graphic, 14 June, 1902. Most of the objects found are preserved in the Greenwich Public Library, and part of the tessellated pavement remains uncovered in situ.]

There is a doubtful record of the discovery of a piece of tessellated pavement in the grounds of Trinity Hospital [Trans. Greenwich Ant. Soc., I, 132].

Shooter's Hill. In 1923, excavations for a hospital on the S. side of the main road and on the top of the hill, revealed the remains of a roughly circular pit or sinking for a former hut. It was about 6 ft. in diameter and 18 in. deep, the floor being covered with a burnt layer with fragments of pottery; a second hut probably 8 or 9 ft. in diameter had been built on the site of the first and at no great distance of time from its destruction. The only datable sherd was of an ordinary 1st-century type [Antiq. Journ., V, 174].