An archaeological watching brief was carried out by Colchester Archaeological Trust on behalf of National Grid during the digging of a trench for a replacement gas main in St Botolph’s Street, Colchester, between TL 9990 2490 and TL9989 2503.<1> The replacement gas main was usually laid alongside and to the east of an existing main. Thus the western side of the trench dug to house the replacement main had typically been dug out before in modern times, while the eastern side was cut through undisturbed deposits. The trench was laid from south to north in St Botolph’s Street, with occasional east-west offshoots to link up with other mains. The trench varied in width and depth. It was typically 0.55m wide, but was up to 1.1m wide in places, and was usually about 1.2m in depth. It was dug by a mini-digger with a toothless bucket. Natural sand and gravel subsoil (L8) was not located in the trench in the southern half of St Botolph’s Street. In the northern half, it was reached in several places in the bottom of the trench. Outside the walled town it was observed at depths of 1.1-1.35m below the modern street-level (Sx 1), while immediately to the north of Short Wyre Street, inside the walled area, it was 0.8m deep. Archaeological remains defined in the watching brief included Roman street metalling and part of the medieval South Gate into the town: Roman street metalling was recorded in or near the bottom of the trench, especially in the northern 60m or so of the street outside the walled town. It was approximately 0.2-0.25m thick and was reached at 0.9-1.2m below the modern street-level. It was seen most clearly outside 5-10 St Botolph’s Street (Sx 1). A disturbed layer of gravel metalling or make-up was observed immediately north of Short Wyre Street inside the walled town. This was up to 0.2m thick and was reached at only 0.5-0.6m below the modern street-level. The metalling (L5) was sealed by a mixed, dark greyish-brown deposit (L4; Sx 1), which was 0.1-0.5m thick. This deposit was found over much of the length of the trench, although it varied in texture. It was typically reached at a depth of about 0.7m below the modern street-level. The deposit L4 is probably a post-Roman topsoil, often termed ‘dark earth’. Approximately 8m south of Short Wyre Street, part of a large foundation (F4) was uncovered in the bottom of the trench. It extended for 1.3m north-south and was 1.05-1.15m below the modern ground-level. It was constructed of large blocks of septaria and some pieces of peg-tile set in a pale brown mortar. It sealed or was cut into natural sand (L8). The foundation was cleaned up but remained unexcavated and undamaged during the pipe-laying. It appeared to survive as two layers or courses of stone, or about 0.2m, deep. The foundation was probably part of the medieval remodelling of the South or St Botolph’s Gate. The peg-tile in F4 would be consistent with a date of around 1400 for the foundation, as there was extensive repair and refurbishment of the town wall at this time. The foundation F4 was sealed by an extensive deposit of robbing or demolition debris (F2; Sx 2b), which was 0.35-0.5m thick. There was no definite evidence for the Roman wall or gate seen in the trench. Another mortared feature (F1) was uncovered approximately 30m to the south of F4. This consisted of pale brown mortar, 0.2m thick, containing occasional fragments of septaria and peg-tile. It was aligned north-south and survived for 1.5m in the eastern section of the trench. It lay 0.85m below the modern street-level. It looked like a shallow stone-and-mortar foundation or plinth rather than a layer of demolition debris.