Roman Road to Winchester

CENTURIATION LIMITES WERE LIGHTLY METALLED

The excavation carried out at Scrubbs Farm over Spring and Summer 2011 showed an unexpectedly complex roadway, not a simple estate road as might be expected of a Roman Centuriation limites. The position on the suspected Centuriation grid is shown in http://www.nehhas.org.uk/wincent.htm. The map below shows the final survey of this road, reported in Ref 1. The line is running down to the valley of the river Itchen, beyond which no further evidence has been found. Much of the evidence is air photo streaks. The excavations was to the east of the map, marked S (SU597 307), in permanent pasture which is believed never to have been ploughed. In this region there is a light air photo line which appears to go through a stunted hedge line (though it is actually large mature trees), which enabled the line's position to be accurately located on the ground. There is a single dark air photo line to the south. The base line was laid out at right angles to the light line, with the 12S marker by the field boundary.

The Composite Section is shown in Diagram 1. To the right of the Diagram in Trenches 1 & 2 the natural strategraphy occurs. There is thin Top Soil, then around 10cms of Subsoil. Then a clay material, with two coloured bands, the lower darker red with a boundary between them showing on the section photos. This material does not pass the clay test of being rolled into a sausage and not breaking when bent over - but on the depth reached is clearly the natural. Two pits or ditches are seen on the section photos. The southern one from 9.7S running out of the trench is likely to be the Roman side ditch showing as the southern Dark line on the air photos - as several re-cut lines and numerous silt lines are visible on the section photos.

Terrace

At 5S the natural stategraphy has been altered. The upper clay band ends, and the clay has been Cut-back to form a level clay surface to the northern end of Trench 4 at 4N. This coincides with the estimated position of the light air photo line. Boundaries are seen on the section photos which suggests that the clay cut out has been re-deposited to form the level surface, shown by R on the Composite Section. The boundary below that has the same slope as the lower clay boundary in Trenches 1 & 2 - and probably is the continuation of that boundary.

Previous excavations have indicated that the Roman practice was to take off the soils, and build their road on the natural. This has occurred here, with the upper clay band taken as part of the soils. As can be seen the land falls slightly to the north. This slope is magnified in the Composite Section, and is hardly noticeable in this field. This road is almost certainly an estate road, not a through Roman Road, but even so the Romans have gone to the trouble and skill to create a level way to put their road on.

The Subsoil above this levelled clay surface is far thicker, lighter, and more homogeneous than in Trenches 1 & 2. This approximately dates the feature, as subsoil takes around 1000 years to form, and so is composed of silting up, and whatever the Romans used as running surfaces for such estate roads. Early in the excavation a band of fine chalk fragments appeared in the subsoil, its distribution is shown on the Composite Section. However it starts around the centre of the levelled clay terrace, where it forms a level band, but continues northwards no longer level - until a massive packed flint agger or curb occurs starting in Trench 6. This feature could not be fully investigated in the time. The chalk band is likely to have been the final re-surfacing, linking the original estate road with probably a later flint structure.

Crushed Flint

As in the excavations on the northern surveyed Centuriation road (Ref 2), a lot of crushed flint came up. Further measures were taken to record it. Such flints were put on the side of the trench, and counted by meter at each break - when the depth of the trench was also recorded. The Histogram in Diagram 1 resulted - which shows higher concentrations than at Roe Down (Ref 2) - while evidence is also available of the depth of occurrence. Such large occurrence is taken as man's activity. No obvious such layer could be seen in section, but flints can be moved upwards by frost action and also dispersed by roots (Ref 3).

The natural occurrence can be expected in Trenches 1 & 2 before the Cut-back, where it averages under 100 crushed flints per cubic m - and almost all came from the Top Soil. It is understandable that Crushed Flint would mainly arise here due to greater frost action and movement of animals and humans. As the Cut-back is approached the Crushed Flint count rises, and averages over 300 per cu m in Trench 3. But this mainly occurs between 30 - 40cms below the Ground Surface - i.e. on the levelled clay surface - and none occurs in the lower natural clay layer. This suggests these Centuriation estate roads or limites were initially given a light metalling of Crushed Flint laid on the levelled clay surface. This may have given traction when the clay was wet. Margary thought such roads had no metalling (Ref 4).

In Trench 4 however the Crushed Flint count rises enormously, and is never less than 1000 per cu m. They mainly occur between Ground Surface and 40 - 45cms below, with none likely to occur in the lower natural clay layer. The interpretation is that this limites was re-surfaced a number of times, but each tended to lay the road further north down the hill. Other occasions are known where this happened, and the initial dead level way for the road was lost (Ref 5). Thus some parts of the road had multiple metallings, while others at the edges had only one. Thus moving into Trench 5, the 1000 figure is maintained initially, but then falls away to virtually nil in Trench 6 - and this maintained over the large flint structure. Most of the Crushed Flints in Trench 5 came from 10 - 30cms below Ground Surface - and none from the lower clay layer.

The highest Crushed Flint count is at the northern edge of the levelled clay surface, which is expected, after which the two clay layers reappear. This indicates that these later re-metallings are just laid on un-engineered ground - though probably the Subsoil was taken off as the chalk is laid directly on the clay in Trench 6 - the subsoil is still thicker now than in Trench 1 & 2. Thus we can see that the final metalling of chalk repeated the process, going further north than the Crushed Flint, up to the packed flint agger - giving evidence that this structure was later. This chalk metalling starts nearly half way along the original engineered terrace, and is 40 Roman feet wide - which together with the pack flint agger is wider than most major Roman Roads.

Ditch with silt lines

There is a very interesting ditch just at the northern edge of the packed flint agger, visible on the section photos, with re-cuts and silt lines. This can be seen on the section photos without any special processing, and can be viewed at http://www.nehass.org.uk/sf11px32.htm. This may be the northern side ditch of the road, which was not seen on the air photos for this field or the one to the east. But occurring right on the edge of the packed flint agger it may be a later feature.

Optical ditch

A students' binocular exercise revealed a large ditch or depression to the north of the excavation, shown in Diagram 2. Binoculars can make slight depressions or swellings stand out. A wide silted up ditch may give this effect, and this one can be seen with a naked eye when one knows its there. The southern edge appears to be the packed flint agger, and the deepest part some 6m to the north. The northern edge is near 20m further north. The extent of the feature to the west of the base line is shown in Diagram 2, and the edges are approximately parallel to the road. The complexity of this limites may be partly due to some installation being constructed by it, and possibly the packed flint agger may be to provide hard standing at its edge.

Acknowledgements

Thanks are due to Mrs Gillian Gray Knight for permission to excavate on her land, and to the 73 man days contributed by the diggers. Richard Whaley

REFERENCES

1. Richard Whaley, Centuriation through Alresford, NEHHAS NL 2008, No 107, 2008; Jn Vol 3 No.16 2012
2. Richard Whaley, NEHAS FAB e News 3, Autumn 2011
3. Richard Whaley, Flints on the Roman Road NEHHAS NL 98, Spring 2006; Excavations on a Centuriation Cross Road NLs 98, 102, 106.
4. I Margary, Roman Roads in Briatain, John Baker London 1955 -73, see section on Ripe.
5. P Fasham Roman Road Winchester - Silchester.., M3 Archaeology 1975, M3 Arch Rescue Committee 1976.

DIAGRAM 1 Composite Section. The origin of the Archaeological Grid is at the centre of the light air photo line's estimated position. 12S is to the right of the Diagram and was by the field boundary. The trench was extended to 16N on the left of the Diagram when the packed flint agger or curb was found late in the excavation. The Trench Numbers are shown below the Ground Scale, each 4m long. Part of Tr 2 was stepped 4m west to avoid tree roots. Below these is shown the estimated position of the light and dark air photo lines. Below the Composite Section, to the bottom of the Diagram, is a histogram of the numbers of Crushed Flints per cubic meter for each meter interval.

Diagram 2 The optical ditch or depression. The person with binoculars viewed from around 60N, and moved a person onto the change of slope or bottom of depression where he put down a flag (X). Then the binocular person moved to the west, and another set of flags put down. The land from the southern ditch edge to the base line appeared flatter, and south of the base line appeared to rise.