at either the town's West Gate, the Lower West Gate to its south, or the South Gate Port Way connected
Calleva Atrebatum (
Silchester) and
Sorbiodunum (
Old Sarum near
Salisbury). Both towns predated the
Roman occupation of Britain, and it is possible that the road is pre-Roman in origin. The name "Port Way" is
Anglo-Saxon in origin, and like other ancient routes with the same name, refers to a road between
market towns. From Calleva Atrebatum, the road continued the south-westerly course of
The Devil's Highway (
Margary route 4a) from Londinium. Both
Ivan Margary and
Thomas Codrington believed the road left the town on its western side; Margary favoured the theory that it connected with the town's Lower West Gate, although it possibly connected with the main West Gate.
Sir Richard Colt Hoare suggested that the road branched off Margary route 42 – the road from Calleva Atrebatum to
Venta Belgarum (
Winchester) – immediately outside the town's South Gate; The
Ordnance Survey's 1911
25 inch to the mile map shows the road to be on a heading congruent with the theory it connected to Calleva Atrebatum's West Gate, although a 1989 article in the
Society for the Promotion of Roman Studies'
Britannia journal shows it leaving the town at the Lower West Gate. Less than from Calleva Atrebatum, Port Way ran across an
Iron Age entrenchment near to where the
1985–87 Silchester Hoard of coins and rings was discovered. The road passed near to (or cut across) the Flex Ditch near Silchester, another Iron Age
earthwork. It continued south-west through
Pamber Forest, towards Cottington Hill near the present-day village of
Hannington. sometimes described as being part of the
Icknield Way. Approximately east of this crossroads was a
mansio, the only significant settlement on the Port Way other than its termini. and the Ordnance Survey's 25 inch to the mile map of 1895 marks it as "ROMAN STATION / Supposed to be VINDOMIS".
Charles Roach Smith wrote that the distance of Vindomis from Calleva Atrebatum given in the
Antonine Itinerary – – did not "materially clash" with the idea that Vindomis was the settlement at this intersection. Despite this,
Francis J. Haverfield wrote in 1915 that "there was no town or village at the crossing; so far as we know, there was not even a house at all". Contrary to Hoare's belief that this was the site of Vindomis, the discovery of the Calleva Atrebatum to
Noviomagus Reginorum (
Chichester) road led to the consensus that the settlement was in the area of present-day
Neatham near
Alton. No later than the 1730s,
John Horsley had suggested that Vindomis was in the vicinity of
Farnham (some from Neatham). If not Vindomis, the settlement at the East Anton crossroads may have been
Leucomagus. , the road entered the town at its Main Gate on the eastern side Quarley Hill provided a line-of-sight with Sorbiodunum, where it was met by Margary route 45a, the road from Venta Belgarum. west-south-west to Durnovaria as Margary route 4e, and finally west to
Moridunum (near
Axminster) and
Isca Dumnoniorum (
Exeter) as Margary route 4f. Margary gave Portway the number 4b, and wrote that it was long. Both Calleva Atrebatum and Sorbiodunum are listed in
Iter XV of the Antonine Itinerary, although the distance given between the two towns – 55
Roman miles – is via Vindomis and Venta Belgarum rather than a straight route along Port Way. == Construction ==