A  Magnetometry survey of the surrounds of the villa and associated test-pitting in the winter of 2009-10 informed excavation strategy in the 2010 season.
Excavation included the successful identification of the site of work by Charles Darwin in 1877.

1:  Magnetometry Project around Cock’s Farm Roman villa, Abinger 2008-10

A project has been set up to investigate the environs of the Romano-British villa at Abinger using the Society’s magnetometer (FM256 Fluxgate Gradiometer). The Surrey Archaeological Research Framework (SARF) had highlighted the lack of evidence for Roman-period rural settlement in Surrey and the Roman Studies Group chose Abinger as a test site for wider investigation. It is a known villa site to which we have been afforded access, together with its rural surroundings. It was hoped that the results would highlight areas of Roman activity around the villa which could be further investigated by test pitting and, if deemed appropriate, by excavation.

A small team of hardy volunteers surveyed the Canterbury field immediately to the north and east of the villa site in 2008-2009 and a number of features were identified. The most exciting were a series of rectangular fields with the east-west ditch running parallel with the spine of the known villa. A rectilinear feature north-east of the villa on a plateau and a kiln-type anomaly to the east of the villa near the eastern edge of the field were identified.

Early in 2010 test-pitting was carried out to relate the magnetometry results with what was physically beneath the soil. Test pits 1-3 were located at the top of the field to explore the rectilinear feature that appeared in both the magnetometry survey and in a resistivity survey carried out by Audrey and David Graham. In all three pits the light grey-brown sandy plough soil went down to a depth of approximately 0.3-0.35m, beneath which the gritty, yellowy-brown subsoil containing varying amounts of ferruginous carstone continued down to about 0.5m before hitting the natural sands (or clay lens as was seen in one pit).  No structures were identified and it was concluded that the anomaly was of a geological nature. Test pit 4 was located over the kiln-like anomaly. Below the plough soil the subsoil was seen to extend down to at least 60cms below ground level, at which point an arc of red burnt sand edged with chalk lumps was identified. Having recorded what was found the test pit was filled in with a view to a more detailed excavation in June. Test pit 5 was placed over a section of the east-west ditch anomaly and almost immediately below the plough soil the darker edge of a linear feature was uncovered. Romano-British pottery sherds were found in the ditch fill and the trench was backfilled to allow for more formal investigation.  A final test pit was opened just to the south-east of the villa over what appeared to be a curving ditch which was thought to relate to the ditch immediately behind the villa (which was excavated in the 1990s).  There appeared to be no sign of the ditch well into the subsoil and torrential rain halted further investigation. As this feature is towards the bottom of the sloping field it is likely that any feature would be some metres under colluvium; it is hoped to investigate this anomaly further at a later stage.


Two fields (Home Meadow and Long Meadow) to the south of the villa and just to the south of the A25 were next to be surveyed with the hope of finding a continuation of the field system. Home Meadow showed a number of wells and areas of burning which are thought to be the result of post-medieval activity relating to Crossways Farm, as well as two modern pipelines. Long Meadow showed evidence for a track up to an old quarry and a modern pipeline but no evidence for an underlying field system. Currently the team has permission from The National Trust to survey East Lawn, the field immediately north of Canterbury Field. Initial results are promising with evidence for a continuation of the Romano-British field system previously identified. Many thanks to Emma Corke, David Calow, Alan Hall and Gillian Lachelin for their participation and support of this project.


2:  Excavation at Cocks Farm Abinger                    


As part of the current programme of work at Abinger coordinated by Nikki Cowlard, a two-week excavation was carried out in June by the Society’s Roman Studies Group. The aims were to continue the search for Darwin’s trench and the Farrer building and to follow up the results of test trenching carried out earlier this year to examine anomalies identified so far by the very informative magnetometer survey (see above). The weather proved to be better than forecast and all aims were met and even exceeded.

Four trenches were opened. Trenches 2 and 5 explored the area thought most likely to include the rooms found in 1877 and there is little doubt that they succeeded. Trench 2 also provided the added bonus that it just clipped a trench dug in the 1995-1997 excavations with the result that this part of the earlier dig can be very accurately related to the current excavations. The upper levels of Trench 2 had been subjected to a great deal of disturbance: there were substantial areas of mostly clean sandy soil fill over pronounced burning below which was disturbed material with Victorian rubbish. Some of this was followed down into a very deep square-cut trench running north-south, which proved to lie alongside the remains of a wall but was cut down well below it into the natural subsoil. The other side of the wall had the remains of what was probably a prepared level for a tessellated floor (the area above it produced many loose tesserae as well as a heavily burnt fragment of an Argonne decorated vessel which can be dated early-mid 4th century but unfortunately cannot be regarded as in a secure archaeological level). It is difficult to see any other interpretation for the deep Victorian trench other than that it is the eastern of the pair dug for Darwin (the one he published: 1888, 182); the floor would be that of the 19th century so-called ‘atrium’. The wall between them met an east-west wall marked by a line of yellow mortar at the base of a robber trench containing much loose mortar and small rubble, which in turn had another wall meeting it on the other side at right angles a little to the east of the wall alongside Darwin’s trench.

Trench 5 was placed 1.5m east of Trench 2, and matched it in having a great deal of 19th century disturbance topped by burning and a clean sandy soil backfill. The east-west wall from Trench 2 was found to continue, but soon met one at right angles forming a T-junction; both were very clearly marked out by yellow mortar lines, with some surviving courses where these could be seen following removal of later disturbance. This junction is not as shown on the 19th century plans (there should be a dog-leg junction) but the area beyond the north-south wall line was full of stone rubble over which a mortared surface survived and it is possible that the Victorian excavators were confused into creating a wall where none existed. West of the wall and surviving only in the section was a clear floor level formed of a layer of packed stone with a whitish mortar topping; it probably matches one seen in the 1990s.

If we accept Darwin’s information then we must conclude that much of the wall material was later removed, probably for use in a nearby garden wall, and that the robber trench was created in the 19th century. It is known that the remains found in 1877 were left open (with a temporary covering) for at least three years, by which time the ‘atrium’ floor was in a poor condition and no doubt material had eroded and/or been pushed with other rubbish into the ‘Darwin trench’ and other deeper cuts. Perhaps after that most of the better stone was removed for reuse in the garden wall, the temporary cover was demolished and burnt in the holes left on the site and the area was levelled up with clean sandy soil brought in from elsewhere. The whole area may have been stabilised by some sort of construction using a line of posts set just to the north of the hedgerow, represented by the post holes noted last year.

The other two trenches were placed in the nearby field and incorporated two of the test pits. The excavation showed how difficult it can be to recognise cut features in sandy soils; even tracing the test pits dug and backfilled only a few months previously proved to be surprisingly difficult.

Trench 3 examined an east-west ditch not far to the north of the villa, part of a field system identified in the magnetometer survey. Large amounts of pottery were found (including pieces of a fine black gloss Moselkeramik beaker), probably dating mostly to the 3rd century (they have not yet been properly examined), and suggesting a date at which the ditch went out of use. A smaller linear feature ran parallel to it on its southern side; no dating material was found but it was stratigraphically earlier than the main ditch fill. It is unlikely to have been a marking out gully as that should have been obliterated by the main ditch. If it was an earlier ditch then the contemporary ground surface must have been seriously lowered by erosion, so it is most likely to have been a palisade or perhaps hedgerow trench. There was a great deal of ironstone in this trench, including some very big pieces, especially along the eastern side. At first this was thought to be the result of field clearance, but it became evident that the stone was also in the natural sand and it was concluded that in this part of the field there is a natural band of broken-up ironstone running approximately north-south. Some was found in the fill of the main ditch and presumably had fallen in over time, possibly in two separate phases as the central level of the fill seemed to be stone free across the width of the trench. The trench was very valuable as it allows us to date the field system and also suggests a major landscape change in the later Roman period when this ditch must have gone out of use. In this respect it can be paralleled for example at Flexford (see this [or previous?] Bulletin), and presumably indicates a change in the way the land was being used.


Trench 4 was placed over a large anomaly thought to be a kiln and it proved in due course to be a limekiln. To judge from the part excavated (probably about a third), this was essentially a circular pit cut deep into the natural sand. The sides were marked by a band of sand burnt a deep purplish red, with occasionally small pieces of red burnt clay and some burnt ironstone although this last was possibly just part of the natural. The pit was also lined with small chalk lumps, partially burnt especially lower down. This chalk lining was encompassed within a feature mostly made out of shaped burnt clay, probably placed green and burnt in situ. Either the kiln originally had a thin lining of chalk and the clay feature was constructed around it or a repair is indicated. It was right in the corner of the trench and so could not be fully examined but seems to have been some kind of flue. The trench probably cut in part along the centre line of the kiln and, at what is likely to have been the centre, half of a circular feature could be examined, marked by a roughly circular stone capping which turned out to be over a pit whose rim was lined with small chalk lumps. The capping was roughly at the same height as the clay feature, but the kiln floor proved to be much deeper.

It seems that the kiln was created by digging out a deep circular bowl-shaped pit about 5m across and packing it with small chalk lumps, possibly starting with just a thin lining. In use it would have been packed with alternating layers of chalk lumps and fuel. The clay feature must be some sort of flue to provide air for the initial firing in the lime-burning process, and the central feature would be the ‘eye’ of the kiln. More work is needed to understand this kiln, not least to give it a date. There was very little associated datable material and none of it can be regarded as certainly dating the feature. The ploughsoil produced a few sherds probably all post-medieval. The fill produced three very small undiagnostic pieces of pottery (but unlikely to be post-medieval), a probably Roman tile fragment from deep in the fill and a flint-tempered sherd from below the stone capping of the eye. A medieval or Roman date is thought most likely for the kiln but further research is needed.


As usual, the excavation owed a great deal to many people. It was carried out by an experienced team from the Roman Studies Group to whom many thanks are due. A larger team was possible this year thanks to a grant from the Society’s Research Committee. Nikki Cowlard set up the excavation with Emma Corke who arranged the necessary permissions and provided on-site facilities. Nikki, Emma and Gillian Lachelin were site supervisors and Alan Hall coordinated recording. Bill Meads again provided metal detector expertise and other assistance. Finds were managed by Margaret Broomfield and members of AARG. Joanna Bird identified some of the finds and David and Audrey Graham provided assistance with the Society’s new total station.

Darwin, C, 1888. The formation of vegetable mould through the action of worms, with observations on their habits