Monument record MDO18604 - Colliton Park, Dorchester; Roman Building I, 'The Roman Town House'

A Late Roman town house was exposed in the NW corner of Colliton Park during excavations in 1937 and 1938. This building now displayed to the public and is known as the ‘Roman Town House’. It had a complex development history, which in its final form comprised two separate ranges of stone buildings, which were never directly interconnecting, aligned E-W and N-S., which underwent a number of phases of development, plus a number of timber structures. The walls were all similar in construction and consisted of mortared roughly-knapped flints laid in herring-bone courses with some use of limestone for quoins and bonding courses. They were generally plastered on both sides and were painted Pompeian red externally. The roofs appeared to have been of hexagonal limestone roof tiles. Eight of the rooms had tessellated floors. The West Range of eight rooms appears to have grown from a nucleus of two rectangular units which were subsequently joined together and elaborated. One room in this range had a hypocaust. One room in the northwest of this range was probably of two storeys. The South Range in its final form comprised a block of five rooms with a corridor along the northern side, based around a nucleus of a rectangular three-roomed unit. A heated room had been added to the west, a corridor to the north, and a kitchen to the east. There was evidence for an L-shaped post-built timber building forming the north and east sides of a courtyard with the stone buildings, and may have included a kitchen at its southern end. Further timber structures lay to the west of the South Range. The L-shaped building was apparently demolished in the 4th century and a cobbled path was laid over the top. Traces of another stone building were found further to the east, which were also sealed by this path. To the north of the South Range was a well, about 1m, which had been deliberately filled up with demolition debris, including the remains of several small stone columns and hexagonal stone bases, perhaps as a closing deposit. The site does not appear to have been built on previously and the earliest phases probably belong to the early 4th century AD. It probably continued in use until the late 4th or early 5th century, though there is no secure dating evidence for the end of the occupation in this building.

A Late Roman town house was exposed in the NW corner of Colliton Park during excavations in 1937 and 1938. Due to the outbreak of World War II, the excavations were never completed or written up. The following account is based on the interim report (1) and the Royal Commission Inventory entry (2). This building has been uncovered and is displayed to the public. Building I had been deeply buried and was in a good state of preservation. The building had a complex development history which has not been fully elucidated to date. It comprised two separate ranges of stone buildings (which were never directly interconnecting) aligned E-W and N-S, and which underwent a number of phases of development, plus a number of timber structures. The house may have begun as a series of small simple three-roomed rectangular stone buildings which were subsequently enlarged and elaborated. The walls were all similar and consisted of mortared roughly-knapped flints laid in herring-bone courses with some use of limestone for quoins and bonding courses. The walls were generally plastered on both sides and were painted Pompeian red externally. The roofs appeared to have been of hexagonal limestone roof tiles. The site does not appear to have been built on previously and the earliest phases probably belong to the early 4th century AD. The West Range appears to have grown from a nucleus of two three-roomed cells, one at the westernmost end of the range (Rooms 16-18) and the other further to the east (Rooms 10, 14). The western part of the range, in its final form comprised a narrow central room or passage (Room 16). A fragmentary mosaic was found close to the west wall, of coarse red and white tesserae in a grid pattern (4). Room 17 to the south was heated by a hypocaust fed by a stokehole added to the east wall. The arrangement of the flues running up the walls suggest the position of a window in the middle of the south wall. The tessellated floor appeared to have been deliberately broken up and consisted a red and white border of swastika meander enclosing concentric rectangles with a central panel perhaps of linked circles of guilloche and stepped triangles enclosing stylised flowers (4). The painted plaster on the south wall had a dark red border on the base and there were many fragments from panels and a floral pattern with a blue flower and green leaf. One fragment had the cursive graffito PATERNVS SCRIPSIT (Paternus wrote this). An earlier pit beneath the floor in the SE corner of the room produced a shale table leg (3). The northern Room 18 also had a tessellated floor in red on a white ground, of which only parts of the border survived, a swastika pattern within chevrons, with a dentil pattern to the east. There was an earlier pit in the NW corner of the room, with a number of flagstones overlying it, one of which was roughly inscribed VAL. The eastern cell in its final form comprised three rooms (10, 14, 15), the existing northern room (15) is an enlargement and replacement of the original north room. Room 10 contained the main entrance to the wing and there was a window in the south wall. It was decorated with painted wall plaster in a rectangular panelled design with curvilinear elements. There was a tessellated floor in dark grey, white, red, yellow, blue-grey and pale grey, with an original central design of twelve octagonal panels which were formed by guilloche borders, with a guilloche or fret-bordered circle which enclosed a rosette pattern. The border was of swastikas between square chequered panels and two-strand guilloche along the southern side. Room 14 lay to the north of Room 10 and was divided from it by a partition. This room also had a coarse tessellated floor, of broad red and grey stripes. The simplicity of this mosaic suggests Room 14 was an anteroom (4). There was a step up into the northern room 15 which was of a different, more massive, construction with two buttresses to the east. There were straight joints with Room 14, indicating that it was an addition. A part of the collapsed wall lay to the east of this room and it is likely that Room 15 was originally two storeys high. It had the remains of a very fine but very fragmentary mosaic, which has been reconstructed by Cosh and Neal as having a central rectangular panel (now entirely lost) surrounded by an inner border of meandering guilloche with rectangular panels midway along each side containing a medallion, of which parts of two survive. These have been identified as Winter and perhaps Spring (or Autumn). The outer border comprises a band of coarse red tesserae closest to the walls with a band of white stepped triangles and then a swastika meander with rectangular panels containing guilloche (4). An infant burial lay outside the north wall. An earlier L-shaped pit was found beneath the floor. Parts of a later rough limestone floor were found in this room, which sealed a number of coins, the latest being an issue of 367-375 of Gratian. A long rectangular room (Room 13) was inserted between the two original three-roomed units to link the two together, with a step down from Room 10 to the east. It contained a tessellated floor with an overall swastika-meander pattern in coarse red and white tesserae with a border of red and white chequers and a red band next to the wall. Over the threshold with Room 16, there was a coarse red and white tessellated mosaic (the pattern of which reflects the mosaic in Room 13), overlain by a later floor of stone flags. A drain at floor level in the north wall of Room 13 lead out to a stone-lined sump (which was also fed by similar drains from Rooms 14 and 18). A later chalk floor was apparently inserted into this room. A coin hoard of AD317 or later was buried near the east door. A small room (Room 8) was added on to the southeast corner of Room 10. It was accessed by a doorway in the SE corner of Room 10. It was provided with a tessellated floor, which survives almost complete. A rough window sill survived in the centre of the eastern room. Fragments of painted plaster in a panelled design in black, blue, brown, green, red, and white were found in this room. The South Range in its final form comprised a block of five rooms with a corridor along the northern side. Its nucleus was a three-roomed unit comprising Rooms 2, 3 and 6, to which a heated room had been added to the east (Room 7), a corridor to the north and a kitchen to the east. Rooms 2 and 6 had opus signinum floors. There was a plastered semi-circular niche recessed into the south wall of Room 2, perhaps for a domestic shrine. The cement floor in Room 6 may have been a late addition as it seals an earlier stoke hole for Room 7 in the southwest corner. A coin of Constans of 341-346 was found beneath the floor. Room 3 had no surviving flooring, but the walls had painted plaster in a green leaf pattern on red. An infant burial was found in the SE corner of Room 3 and another in the SW corner of Room 6. The corridor (Room 4) ran along the north side, with a step up to the east to form a small lobby (Room 1) at the north end of Room 3. The corridor was provided with a chalk floor. Room 5 was added on to the east of Rooms 1 and 3. Its walls were apparently unplastered and it had a rough limestone floor. In the NE corner of the room was a stone-lined oven and another in the southern half of the room. A rough hearth of tile and stone roof tiles was found near the north wall. This room was probably used as a kitchen. Several coins, the latest dating to 341-6, give a terminus post quem for the limestone floor. Room 7 was added on to the east end of the range, to the east of Room 6 and had a channelled hypocaust with an opus signinum floor supported on large limestone slabs. The hypocaust was originally fed by a stokehole in the SW corner of Room 6, but was later remodelled with a new external stokehole added on to the SW corner of the room. Painted plaster in a panelled, perhaps foliate, design in white and red with green and brown lines was recovered from this room. To the west of the South Range parallel lines of postholes, with some evidence for the replacement of posts, indicated a timber building in this area, perhaps to shelter the stokeholes feeding Rooms 7 and 17, or forming a linking structure between the West and South Range. An L-shaped post-built timber building of two rooms was found forming the north and east sides of a courtyard with the West and South Ranges. The southern room had an oven in its southern end and it may have been a kitchen. This building was apparently demolished in the 4th century and a cobbled path was laid over the top. Traces of another stone building were found further to the east, which were also sealed by this path. To the north of the South Range was a well measuring about 1m in diameter and 10m deep. It was deliberately filled up with demolition debris, including flints, mortar, plaster and stone roof tile, together with eight small stone columns, two hexagonal stone bases, and several slabs. These may have been deliberately placed as a closing deposit. The dating evidence from the building suggests that it was constructed in the early 4th century and probably continued in use until the late 4th or early 5th century, though there is no secure dating evidence for the end of the occupation in this building.

Record last edited

Mar 10 2024 5:28PM