Monument record MDO6398 - Roman temple, Witchampton

A circular building of Roman date excavated in 1923-24. It was around seventeen feet in diameter, and had flint wall footings, and slabs of Purbeck stone formed the remains of a floor thought to have had a mosaic. A rectangular building was attached on the south west, and other buildings were recorded in the vicinity. The remains of one building to the south west incorporated an oven, and flue tiles were also found in this area. Finds of roof tiles, fragments of painted wall plaster and glass indicated that the circular building was a building of considerable status. It is thought that it was most probably a temple with ancillary buildings. Because the site is in a low-lying situation, it has been suggested that it may be associated with the cult of a water deity.

The excavations of a building in Abbey Mead, Witchampton were conducted by Mrs MacGeagh in 1924. The plan consists of a circular structure joined to a rectangular one. The circular structure is built of flint and mortar and remains to only 1 ft 5 ins high; the rectangular structure was built of smaller flints and the foundations are not as strong. A semi-circular gravel concrete floor remained on the west side. Within the circular wall the floor was covered with large slabs of Purbeck stone, with broken tiles, coloured wall plaster, hard mortar with tile in it, a very large tile tessera, bronze rings and a coin of Constantine I but no potsherds. Much the same was found in the rectangular building except a coin of Valens was uncovered. Trial trenches to the SW revealed much fallen building material of Roman date, and an isolated oven or kiln with a stoke hole and stone hearth. Potsherds were found in this area and consisted mainly of New Forest Ware but there was some samian. Two coins of Gallienus, flue tiles, charcoal, oyster shells, animal bones and some iron were also uncovered. Above the Roman buildings were skeletons. 'It is difficult to explain the use of the circular building; possibly it was a funeral monument, possibly it was a domestic building, such as a pigeon house'. <1>

Possible explanation of circular building and annexe is a small temple. <2>

A well built rectangular flint building of uncertain date was uncovered forty feet south of the earlier structure. No Roman objects were found but a farthing of Henry II, medieval pottery and chessmen were discovered on the rammed chalk floor. <3>

The chessmen belong to the period c 1000 to 1100 AD. They are preserved by Mrs MacGeagh. <4>

Norman and English pottery found at Witchampton in 1939 is now in Poole Museum. <5>

No re-excavation has taken place at the site and the area has now been planted as an orchard. No further finds have been made, but the chessmen are now in the British Museum. <6.1>

Other finds were made between 1925 and 1939 in the nature of pottery, coins and some skeletons with no accompanying grave goods. The excavations were made during spare moments from the duty of head gardener to Mrs MacGeagh and upon her death in 1939 the pottery was sent to Poole Museum. 'A long necked vessel and a mortar were found at the edge of my garden in 1935 at ST 99030670. No other finds were made. Vessel owned by Mrs Martin'. <6.2>

Nothing is visible at the site which is grass covered. From Bevin's description the Md building is sited with reasonable accuracy. The siting of the vessel is suspected by the investigator as being false; the objects were found in peculiar and romanticised circumstances and it seems possible that they were found at or near the Md building site and later 'salted' in the cottage garden. The relics are in the possession of Mrs Martin. Mr Bevin was not able to state precisely where the skeletons were found, 'but excavations seem to have been unscientific; and a re-excavation a matter of necessity'.<6>

'At Witchampton (Dorset) an elaborate round mausoleum appeared'. <8>

The most probable interpretation of the circular building is as a temple with an added annexe, as at Pagans Hill, Chew Stoke, Somerset, or at Frilford, Oxfordshire. The cult of a water deity would explain the low-lying situation. The rectangular building from which the chessmen were recovered was probably Medieval, possibly part of the same complex as the old Manor House. <9>

Listed as the possible site of a Roman villa and a temple. <10>

<1> Society for the Promotion of Roman Studies, 1924, Journal of Roman Studies 14, 235-7 (Serial). SDO16808.

<2> Dorset Natural History and Archaeological Society, 1925, Proceedings of the Dorset Natural History and Antiquarian Field Club for 1924-1925, XXXVII (Serial). SDO30.

<3> Society for the Promotion of Roman Studies, 1925, Journal of Roman Studies 15, 238 (Serial). SDO16809.

<4> Dorset Natural History and Archaeological Society, 1926, Proceedings of the Dorset Natural History and Archaeological Society for 1926, LXL, LXI (Serial). SDO20395.

<5> 1949, The Archaeological Newsletter 2 No 6, 100 (Serial). SDO20392.

<6> Quinnell, N V, Various, Field Investigators Comments NVQ, F1 NVQ 29-MAR-54 (Unpublished document). SDO11903.

<6.1> Woodruffe, Miss, Oral: Miss Woodruffe (tenant), Abbey House, Witchampton (Verbal communication). SDO20394.

<6.2> Bevin, Mr, Oral: Mr Bevin, Lois Cottage, Witchampton (Verbal communication). SDO20393.

<7> Ordnance Survey, Ordnance Survey Map 6in, 1963 (Map). SWX1540.

<8> Collingwood R G and Richmond I, 1971, Archaeology of Roman Britain, 171 (Monograph). SDO17919.

<9> Royal Commission on Historical Monuments (England), 1975, An Inventory of Historical Monuments in the County of Dorset, Volume V (East), 110 (Monograph). SDO129.

'(23) Roman Building (99060641), perhaps a temple, was excavated in the gardens of the Manor House in 1923–4. The site is 100 yds. N.W. of the R. Allen, on a slight hillock of clay on river gravel.
Flint and mortar footings 4½ ft. wide formed a circle 17 ft. in diameter. Upon this were the remains of a flint wall, enclosing a floor of large Purbeck stone slabs; apparently these had carried a mosaic floor set in mortar. On a platform of compacted gravel to the S.W. were the slighter flint footings of an attached rectangular room (21½ ft. by 50 ft.), divided by a wall 9 ft. from the N. W. end. A semicircular extension of the gravel platform, 35 ft. wide, projected 14 ft. further S.W. Roofing tiles, painted wall plaster, glass fragments and a coin of Constantine I lay on the floor of the building. To the S.W., more debris and an oven indicated another building. Flue tiles, and coins of Gallienus, Constantine and Valens were found in this area.
The most probable interpretation of the circular building is as a temple with an added annexe, as at Pagans Hill, Chew Stoke, Somerset, or at Frilford, Oxfordshire. The cult of a water deity would explain the low-lying situation. The finds are in Poole Museum. (J.R.S., XII (1922), 268; XIV (1924), 235–7; XV (1925), 238. Notes in D.C.M.)'

<10> Scott, E, 1993, A Gazetteer of Roman villas in Britain, 55 No 1 (Bibliographic reference). SWX1795.

<11> Historic England, Historic England Archive, BB72/03972 (Index). SDO14738.

PROBABLE TEMPLE RESEMBLING CHEW STOKE,SOMERSET.SEE JRS/14 (1924) 235-7

<12> Historic England, Historic England Archive, BB72/03973 (Index). SDO14738.

MD BUILDING TO NW OF ROMAN TEMPLE RCHM MON 21

<13> Historic England, Historic England Archive, BB72/03974 (Index). SDO14738.

PROBABLE TEMPLE RESEMBLING CHEW STOKE,SOMERSET.SEE JRS/14

<14> Historic England, Historic England Archive, BB72/03975 (Index). SDO14738.

PROBABLE TEMPLE RESEMBLING CHEW STOKE,SOMERSET.SEE JRS/14

<15> National Record of the Historic Environment, 209265 (Digital archive). SDO14739.

Record last edited

Dec 21 2023 12:28PM