While there is evidence of organic town development in the first decades of Roman occupation (c.AD 50–c.90), the end of the century may have seen the planned building of something like a religious ‘resort’. Very similar to establishments are known in Roman Gaul (modern France and Belgium, which had been incorporated into the Roman Empire nearly a century before the Roman invasion of southern Britain). It might even have been entrepreneurial, immigrant Roman Gauls who established a shrine or shrines beside the Stour. Such resorts were typically equipped with a theatre and public baths at an early stage. Certainly, local people were involved in this development; the native peoples (now known as Romano-British) were already beginning to create their own particular brand of ‘Roman-ness’.  The first major change to the town took place in the early second century, c.AD 110, with the establishment of a new street pattern. The main streets were laid out at regular intervals, probably using the Rutupiae road as a base line, with a street between baths and theatre, and another street linking with the main road from Dubris (Dover), which was becoming an increasingly important port. Roman Canterbury perhaps only now boasted a forum and basilica and the status of a civitas capital — the administrative centre of various British groups, collectively known as Cantiaci by the Romans. During the second century, the early cemetery at Rosemary Lane, was superseded, and ribbon development spread along the main roads out of the town. Within the town, new cross-streets were built at right-angles to the new axis. The road north-east towards Regulbium (Reculver) may also at this time have been laid out on a more perpendicular alignment to meet the Rutupiae road. At the beginning of the third century, Durovernum Cantiacorum was graced with an imposing new theatre in classical style, adding to the monumental appearance of the temple precinct. Perhaps only then was the temple rebuilt, also in the classical style. It’s possible that a residential ‘north-east quarter’ of the town only developed at around the same time in the area of the later cathedral. This seems, in fact, to have been the area that was to become the Anglo-Saxon inner ‘burh’ and medieval priory and cathedral church precinct. It also, perhaps, had a more typically Roman layout of streets on a grid pattern. Two Roman-period streets beneath the cathedral are known from archaeological evidence. Other evidence might point to further lost Roman streets in this quarter: for example, the boundary of the Anglo-Saxon ‘bertha’ demarcating the Liberty of Christ Church, and the medieval street lost to Archbishop Lanfranc’s take-over of land for his new palace (between The Borough and Orange Street). Another Roman street is known at St Radigund Street, and this, and (probably) all those within what became the cathedral precinct, met with a street aligned north-west to south-east which may have marked the north-east side of the town proper and informed later development. In the late third century, the town was extensively remodelled and reshaped by an enclosing wall, rampart and ditch, producing the bounded shape familiar today. The Roman walls became the medieval walls shown on the map. The new boundary actually had the effect of cutting off not only a ‘north-west quarter’ in the St Dunstan’s area but also earlier ribbon development along other major roads leading away from the gates. Later cemeteries developed closer to the new perimeter wall, in areas by then considered to be outside the town.