near the
Newstead Helmet The fort was constructed in multiple phases. Dr Simon Clarke of
Bradford University has produced a logical sequence of building and destruction for the fort and its annexes. This was managed by combining evidence from the first excavations of James Curle and Sir
Ian Richmond with aerial photographs and modern search and rescue excavations of Bradford University (1987–1997)
Phase 1 (c.79-87 AD): The earliest occupation of the site identified by aerial photographs since the 1940s have revealed at least three and possibly as many as five camps in this area. The enormous "Great Camp" identified by Curle was on the highest point of the ridge at the centre of the whole enclosure, commanding a view over the road from the Cheviots, and across the Leader valley to control the crossing of the river. It had an area of and each of its 4 gates had a
titulus, a long straight ditch for defence, dug in front of it. Its size means it must have been constructed for a large, probably legionary, force and its occupation must have been of short duration. In Scotland the only excavated camp that compares with it in size is the legionary fort at
Inchtuthil of . In Wales,
Caerleon of occupied by the Second Legion, corresponds closely in area. The later smaller
Agricolan fort was built to the west of the Great Camp about 80 AD by the
Legio XX Valeria Victrix mentioned on many of the inscriptions. It had a turf rampart on a cobble foundation with two ditches in front of it, overlapping each entrance. On the west side was an annexe which was also defended by a similar rampart and ditches arrangement.
Phase 2 (c.90-105 AD): After a possible short abandonment of the fort, the Romans were back, and building in strength. Old ditches were filled in and new defences constructed. This resulted in a colossal strengthening of the fort. The new turf rampart was built on a cobble base which measured 13.5 m across and around 8.4 m high. In front of this was a single ditch between 5 & 7 m wide and 2 to 4 m deep. New, well defended annexes appear on the south, east and probably north sides of the fort, inhabited by civilians and camp followers.
Phase 6 (c.160 AD): Around this time the previous construction of the subdividing wall was removed as Trimontium's role changed from supply and manufacture to a front line fort due in part to the abandonment of the
Antonine Wall. Within the fort a long, narrow barrack block was constructed and evidence points to a large decrease in the civilian population surrounding the fort. that can be seen at the
National Museum of Scotland in Edinburgh, and intriguingly the prospect of a gyrus, or training ring. During these phases of occupation, the population of the fort varied considerably, the permanent garrison level was probably around 1000 but this number would be complemented by the trades, manufacturers, craftsmen and families associated with the camp during the different phases of occupation. It has been estimated that the number could have risen to anywhere between 2000–5000. ==Site archaeology==