Roman Road to Winchester


The wide ditch, found to be the source of a dark air photo streak running from high ground into Chawton Park Wood which was excavated over 2006 - 9, was completed in 2010. A map of the excavation site on the western edge of the Wood is given in, which gives a revision of features in this area (Ref 1). The Composite Section for the 2010 work is:

Key to Composite Section: L is Light Subsoil as seen in the ditch and over the northern running lane. RW are Rabbit warren tunnels. AP - D shows the position of Dark Air Photo line and - L the position of the Light line. C is crushed flint. Non-hatched layers are natural clay. You can print out the rest of the ditch from Ref 2, setting Landscape and margins 18mm. This will fit together with the printed Composite Section above.

The dark air photo line is bounded by light lines, with an intermittent light line in its centre - which has not been seen in excavation. The Composite Section above shows similar arrangement as the northern side of the ditch in Ref 2 from 13 - 20S. The natural clay, which has been cut down 30 - 40cms (and silted up with a number of layers of subsoil), rises rapidly from 41S to reach its more usual depth around 43S. The clay however continues to rise, becoming near the Ground Surface at 46S. This parallels what was seen from 20S to 16S, the interpretation being that clay from the ditch was piled up on the edges - giving buried Top Soil visible 15.5 -17S. The nearest southern equivalent is the light coloured clay layer of increasing thickness from 43 - 46S. The lower boundary of this layer has a Top Soil coloured tinge on the section photos. By 48S the clay is back to its usual depth.

Some of these features are masked by a rabbit warren which we were advised had been taken over by foxes, who typically build up a thick layer of Top Soil through cleaning out their tunnels. Initially thinking it was a badger set we did not intend to excavate, but key information was contained within it (designated as Trench 13 to 14, 40 - 43S).

Crushed Flint Running Lane

The northern side of the ditch had a packed flint running lane built into the side of the ditch as the clay is cut down. No such structure was found on the southern edge. It was noticed that a lot of crushed flint came up, as had been seen on other digs, interpreted as a top dressing which had got washed out by water action. Flint occurs as quite large rounded structures, and if a lot is crushed into small stones it may well be by man's action. A lot of discussion has occurred in this project of the movement of stones in clay environment (Ref 3), causing an ordered layer to be randomised and dispersed. With smaller stones or fine material used as a running surface dressing, roots can penetrate on abandonment, which will eventually turn to soil. A combination of these effects may mean that such surface dressing is not noticed on trowelling, or not noticed until a body of large foundation stones is reached - causing the surface dressing to be lost. Methods were devised on this excavation to establish where the crushed flint came from.

Crushed flint was seen in Trench 13, 36 - 40S. The limits and intensity of this crushed flint was found by trowelling down a few centimetres, recovering the crushed flint, laying them on the new surface, and making a sketch of them. Abruptly at 29cms below Ground Surface the replaced crushed flints occupied much of the space they had been taken from. This surface is represented on the Composite Section by the dashed line from 39 - 40S. This continued down to 43cms below Ground Surface, when abruptly the flints were mainly unbroken - below which there were no flints to the Trench Bottom. Here there was clay which nearly passed a clay test.

This bottom of flint level shows up on the section photos as a double boundary, looking as if they are impressions of larger flints - shown as a double boundary of the Composite Section. No boundary is seen on the photos of the crushed flint surface represented by the dashed line - though a boundary with the light subsoil takes over from it on either side. What is interesting is that this double flint bottom boundary rises up on the section photos to meet the light subsoil boundary - for this is the same behaviour as the northern running lane which tapers to a point at 20S. Further, these two tapered points are exactly half a Roman Actus apart (59.99405 Roman feet). This is pretty conclusive evidence that we have Roman structures, and that we have a southern running lane which was lightly metalled with crushed flint.

The double boundary continues into Trench 13 to 14, and starts to rise up at 41S. Projecting where this meets the light subsoil surface gives the same width as the northern running lanes (12.2 Roman feet - about half the common width of Roman Roads). The height of the southern running lane is a few centimetres above the northern - but if the northern had a surface dressing of crushed flint which we did not see then the two running lanes could be at the same height.

Crushed flint was gathered and bagged with details of where it came from, which leads to the Stone Count per cubic meter diagram below the Composite Section. This shows a similar peak stone count over the running lane as seen for the northern lane (though here the count was not specifically crushed flint and the flints tended to be bigger). The sharp fall in stone count for 40 - 41S, where the running lane is still occurring, is due to the deep Box Section cut here to reach the clay and the very thick Top Soil. The stone count is expressed in cubic meters of trench excavated. There is information of occurrence of crushed flint with depth, and none occurred at the deeper levels or in the Top Soil, but it is not possible to adjust this plot, which needs to be consistent for the whole excavation. If however Trench 13 to 14 had been only as deep as Trench 13 (which would also mean less top soil) then the Stone Count would be 500 flints per Cu m for 40 - 41S.

Swelling & Depression in field

Two swellings over the running lanes and re-deposited clay, with a depression between them representing the silted up ditch, have been seen in the drill lines of young crops and captured in a photo: In Easter 2011 they were clearly visible for some way into the field, and the farmer gave permission to survey them, and took part himself with his binoculars. Binoculars magnify the impression of the swellings and depression, and Alex was moved onto the summit of each swelling, where he put a marker. This was repeated until the northern swelling became indistinct. The southern swelling ran into another feature, where the survey ended. The northern swelling was a pretty straight line of markers, and its angle was measured at the Base Line at 18.9S with 98degrees. This is over the running lane. The Southern Swelling was a more wavy line, its average cutting the Base Line at 43.25S with 97.5 degrees. This was over the re-deposited clay. The Northern Swelling could be detected for 77m from the Base Line, and continuing this line cut Roe Downs House as the light air photo line. The Southern Swelling was detected for 73m. Clearly this ditch feature is the same as the air photo evidence. While the central light line of this dark air photo streak has not yet been seen in excavation, it does show up in the above photo as a small swelling at the bottom of the depression - which could well be a central running lane. This would add further complexity to this feature - but does not add yet to the interpretations given in Ref 1.


Thanks are due to Mr John Cray for permission to excavate on this land and for backfilling; and to the 93 man days contributed by the diggers. Cristina Hellers compiled the Composite Section. Richard Whaley

1. Richard Whaley, Chawton Park Wood Zig-Zag starts Centuriation, FAB e News 3 Autumn 2011
2. Richard Whaley, Possible 2 or 3 lane highway found on edge of Chawton park Wood, FAB e News 1 Autumn 2010
3. Richard Whaley, Flints on the Roman Road NEHHAS NL 98, Spring 2006; Excavations on a Centuriation Cross Road NLs 98, 102, 106.