Page:VCH Suffolk 1.djvu/361

 ROMANO-BRITISH SUFFOLK and Dunwich, making it run by Coddenham, East Soham, and by Sibton Abbey, and so to Dunwich. There is one objection to this route, that it passes by no site with so many Roman remains as at Burgh, which might well be taken for the Combretonium of the Itinerary." No line of commu- nication has yet been discovered between Walton Castle and the routes here mentioned. It is impossible that there should not have been such a line. As for Gariannonum (Burgh Castle), a road starts from that fortress south- wards, which is supposed to have passed by Hopton and towards the end of Oulton Broad. It cannot be traced farther, if as far. Possibly the road continued till it joined the line from Dunwich to Wangford; or perhaps, crossing the Waveney at Burgh St. Peter, it may have passed on to join this same line near Ditchingham station in Norfolk; but this is mere conjecture.*' The most direct road from the south to Venta comes from Stratford St. Mary (Ad Ansam), and as far as the Gipping is the same road as that spoken of as part of Iter IX. From thence, proceeding in a nearly direct northern course, it runs by the Stonhams to Stoke Ash, and so to Scole, where it passes into Norfolk on its way to Venta Icinorum. Though by far the most direct route to the last-named station, it cannot be identified as belonging to the Antonine system. ' The Peddars Way,' one of the most remarkable of the Roman roads on the eastern side of Britain, is likewise omitted from the Antonine Itineraries. Its supposed starting-point was Colchester, but for i6 miles through Suffolk there are no signs of it. Not until the village of Hitcham is passed do traces of it appear. From Hitcham it goes on to Woolpit, from thence to Stowlangtoft, and so on past Thetford, where crossing the county boundary about 5 miles east of that town it enters Norfolk. It crosses the west side of Norfolk to find an ending at an obscure village. Holm on the Sea, at a point where the Wash is narrowest from the opposite coast of Lincolnshire, and 4 miles west of Brandon (Branodunum), the first of the forts of the Saxon shore. A writer, previously cited, says of it, from Stanton ' this remarkable road can be traced for 45 miles to the north coast of Norfolk. It has been called British, but it has all the charac- teristics of Roman laying out, and is, indeed, the best-preserved Roman road in East Anglia.' And, again, Peddars Way has been said to pass through neither town nor village, and it is true that from Ringstead (in Norfolk) 'southwards as far as it can be traced, about 45 miles. Castle Acre' (a possible Roman station) ' is the only village upon it, and those near it are small.'*' One other road requires to be mentioned, viz., the Icknield Way, which shows faintly on the western border of Suffolk, passing the county boundary at Kentford, and running on less and less visibly towards Lackford, but vanishing beyond that place. " T. Codrington, Roman Roads in Britain, 220-1. " Ibid. 222. The writer apparently here confuses a place on the Waveney called Burgh St. Peter with the great fortress at that river's mouth, or he may have been led into error by the vain imaginings of local antiquaries, one of whom at least has endeavoured to fix the site of Gariannonum on this spot in order to bolster up a theory, unfortunately too much credited, of the existence in Roman times of a huge estuary separating Norfolk from Suffolk and running far up into the former county. There are no signs of any Roman fortress at Burgh St. Peter, nor have any Roman remains been found there. 299
 * T. Codrington, Roman Roads in Britain, 224 et seq.