Roman Road Winchester to London

Solution to the three parallel features in Chawton Park Wood

We have been excavating the Roman Road on the western edge of Chawton Park Wood since 2006. It is believed to be the Roman Road from Winchester to London, which is also part of a Roman Centuriation - where a Roman estate is provided with a grid of parallel and right angle Roman estate Roads. We have found three parallel features in the Wood. OS map error has resulted in inaccurate plans if the features are added to OS maps, so we are making our own maps from Ref 1 Air Photos. This region is also likely to be the start of the Roman Centuriation estate (Ref 2), but at present we do not know which of these three features forms the Limites or estate road and boundary - which has been established some 2 miles to the west on towards Winchester. Using Air Photo Maps has solved where the features go around the excavation site (Ref 3), and where they start in Chawton Park and Ackender Woods (Ref 4). This report charts where these features go westwards, and connect with the Centuriation surveyed on towards Winchester.

Above is the Air Photo Map which bridges the gap between the known features around Chawton Park Wood on the East, and the previous published report on the West (Ref 5) - this is reproduced in the Air Photo Map at the end of this report. The excavation sites at Roe Downs Farm are shown in the next Air Photo Map to the East marked S. The excavation on the Southern Feature found a Roman Road metalled with crushed flints, with packed flint revetments on either side (Ref 6). On the Northern Feature a Road was found with unusually deep and wide ditches, and probable revetments (Ref 7). Excavations on the Central Feature found a 60 Roman foot wide ditch with metalled running lanes on the ditch edges (Ref 8). Ref 4 gives an interpretation of a water feature to collect water for a mill at the bottom of the valley, and possibly for hauling logs down the valley side.

In 2005 the route from the western side of the above Air Photo Map could not be continued beyond the Ref 5 publication. The gap was not large, but the line ran into hedgerows, then into an area of development (where we still have little evidence) - and the line did not marry up with the evidence on the East side of the above Air Photo Map when plotted on OS maps. Indeed it appeared to be nearer the Zig-Zag than the Southern Feature (Ref 9). We now know this was due to OS map errors, but led us to the Central Feature.

On the east of the above Air Photo Map the previously published line of the Southern Feature is shown by the computer generated representation of the air photo streaks - light line bounded by dark lines. It is more accurate if what you want to plot on an Air Photo Map is visible on that air photo, rather than transfer it from another air photo. Extensive coverage of coloured modern air photos was obtained over the area of the above Air Photo Map, but little new information was obtained. From the known easterly route of the Southern Feature, it crosses the modern lane and enters a wood. Here there may be remains of a terrace. The line would pass the southern part of the house to the west of this wood. Here in the track leading south from this house is binocular evidence of two ditches - likely to be the side ditches of the Roman Road (not plotted on the above Air Photo Map). Nothing is visible across the next field, either on the Ref 1 photo though free of development in 1947, nor visible in the field itself. In the next two small fields, the northern field boundary lines up well with the northern air photo dark line and is likely to be following the Roman side ditch. Just to the south of this boundary starts a very narrow light line, shown by a series of small white dots, which can be traced through the hedgerows which defeated us is 2005, and off to the western edge of this air photo (View obliquely into light, with magnifier). We are now on the published output of Ref 5. This narrow light line is probably from one of the revetments found in the excavation. Such a narrow line was reported in Ref 4 going eastwards through Chawton Park Wood.

This line has been traced westward and points accurately at the spot where the modern A31 coming east out of Winchester changes direction more to the north. This A31 has long been thought to be Roman, though evidence for this is not large (Ref 10) - what with long use, A31 construction and dualling. We can now see from the Air Photo Map that the Southern Parallel Feature is part of this line, is likely to be the Roman Road Winchester to London, and from its connection with the A31 likely to be the spine Road for the construction of the Centuriation grid of Roman estate Roads. However in Ref 4 we concluded that through Chawton Park Wood it was a First Road later replaced by the Northern route with shallow gradients. We will return to this.

Northern Parallel Feature

We will now trace the northern feature westward, which luckily is also visible on the Ref 1 air photo obtained to bridge the gap, as a thin light line. Again this is likely to result from revetments. Starting on the east side of the above Air Photo Map with the computer print of the last published course, this is continued through part of the strip fields. On the west side of the modern lane the light thin line is visible through the wood, through part of the next field up to the houses and gardens. After the houses the thin light line continues across the field to a light spot on the air photo, where it changes direction more to the south. Nothing can be seen in these fields, except there is a depression in the region of the light dot. This is likely to have been some sort of quarry, and quite possibly a reason for the Roads construction.

In previous maps of these features they appear to be parallel - but in the above Air Photo Map they are getting closer together. Measurements indicate this is only occurring over the last two fields before the light dot and change of direction. The change of direction means they will meet.


Air Photo Map at the end of this report shows the previously published course from Ref 5 with the computer generated representation of air photo streaks. This air photo overlaps with the one at the start of this report. The thin light line from the Northern Feature can be traced on this air photo, and is shown below as a series of fine white dots. The computer generated air photo streaks for the previous published course are omitted in the field where the two routes cross.

Importantly this Northern route continues after it crosses the previously published Ref 5 course of the Southern route - and heads for the prominent light air photo line on the Ref 1 air photo - which has been enhanced so it is visible on the Air Photo Map below. This feature is prominent on the ground, and was surveyed and discussed in Ref 5. On the Air Photo Map it is significantly further away from the computer generated air photo line of the Southern route than in the Ref 5 survey - illustrating again that it is not possible to make accurate Roman Road maps by adding them to OS maps. This feature was reported by the famous archaeologist A J Clarke, who died around the time of our Ref 5 survey, in private correspondence on his finding this Roman Road from Winchester. We found that this feature was near the air photo route but not on it - and might be the result of map error - but it did not line up with our findings going westward. We can now see it is part of a different Roman Road. A J Clarke appears to have dropped his work on this Winchester Road at this point. Not only did he face map error which would have prevented him from lining up his features with ones in Chawton Park and Ackender Woods, but he had quite a complicated set of Roman features to sort out.

In conclusion, it is the Southern feature which constitutes the Limites of the Centuriation, starting in Ackender Wood (Ref 4). However for much of the way till the Air Photo Map below it was superseded by the Northern feature as the main Roman Road with shallower gradients over difficult country. The construction of the Northern route at the excavation site on Chawton Park Farm was seen to be superior to that on the Southern route at the edge of Chawton Park Wood by being of packed flint. The Northern route was also seen as packed flint on the Zig Zag. But on the edge of Chawton Park Wood both routes were of crushed flint, with packed flint revetments.

Principal Surveyors

Richard Whaley

Map Caption The first Air Photo Map made from Ref 1 Air Photo 41/63 NE from Chawton Park Wood to Soldridge (SW of the Photo). The 2" OS Grid Bars represent 1760'. The second Air Photo Map is made from Air Photo 41/63 NW Soldridge to Stancomb Farm (lower centre of picture)


1. RAF air photo survey 1947-9, scaled to 6" : 1 mile by OS, Hants Sheet 41/63 NE & NW

2. Note on Centuriation, NEHHAS NL No.91 Spring 2004, Journal 3 No.10 2007

3. Richard Whaley, Chawton Park Wood Air Photo Map, NEHHAS FAB e News 4 Spring 2012

4. Richard Whaley, Southern Parallel feature in Chawton Park Wood , NEHHAS FAB e News 10 Winter 2014-15

5. Centuriation west of Four Marks, NEHHAS NL No. 98 Spring 2006, Journal 3 No13 2007

6. Excavations on the Southern Parallel Feature, FAB e News 15 Autumn 2016

7. Excavations on the Northern Parallel Feature, FAB e News 15 Autumn 2016

8. Excavations on the Central Feature FAB e News No. 3 Autumn 2011

9. Chawton Park Wood Zig Zag, FAB e News No. 5. Autumn 2012

10. A31 East out of Winchester, Roman Road Abstracts, NEHHAS FAB 2016