A watching brief was undertaken by Colchester Archaeological Trust at the site of the former CBC depot at Jarmin Road in Colchester during the machine-excavation of a series of test-pits by the developer. <1> Thirteen test-pits (TP1, TP3-TP11, TP13-TP15) were dug using a 360° mechanical excavator. A couple of other test-pits (TP2, TP12) were abandoned and several test-pits had to be realigned slightly or extended due to the presence of services or intractable modern obstacles. At the time of the watching brief, all of the buildings on the site had been demolished, although their footprints could still be made out. The test-pits varied in size, but were usually 2.5-3m long x 0.75m wide and roughly 3m deep. The natural subsoil (L5) was reached in all the test-pits. The depth it was reached at varied from 0.5m below the modern ground-level in TP1 to just over 1m in TP9 and TP10. Two probable Roman features were revealed in the southwestern part of the site. There was a spread of fragments of burnt daub and Roman brick/tile across the site. These probably derive from a Roman tile kiln on or near the site. The natural subsoil was contaminated in places by leakages from underground fuel tanks. A shallow Roman feature (F1) was revealed in TP9 in the south-western corner of the site. It had a dark greyish-brown fill and extended over the northern 1.35m of the test-pit. It was approximately 0.55m deep and appeared to be either a pit or part of an east-west ditch. A small quantity of Roman pottery and two fragments of Roman tile were recovered from the backfill of F1. The sherds dated from the mid 2nd to the 3rd century. This feature lay approximately 2.5 m south of the east-west Roman ditch (F19) uncovered in T3 during the 2007 evaluation (CAT Report 409, 2; <2>). Test-pit TP10 cut across the western end of 2007 evaluation trench T4. A shallow east-west ditch (F2), with a yellowish-brown fill, was exposed in TP10. This was the same feature as F16 in T4 in 2007. The ditch was approximately 2m wide and 0.5m deep. No finds were recovered from the ditch in either 2007 or 2008. It was perhaps a continuation of ditch F19 uncovered in T3 in 2007, and thus was probably Roman in date. A brownish layer (L4), approximately 0.2 m thick, was observed in many of the test-pits. This layer was probably largely re-worked subsoil. It sealed the natural subsoil (L5) and was sealed by post-Roman topsoil (L3). Many fragments of burnt daub and Roman brick/tile were observed in L4 and a sample was recovered. Some of the fragments of Roman brick/tile were possibly wasters. The frequency of the burnt daub and brick/tile fragments varied from test-pit to test-pit, although they were particularly marked in TP3, TP4, TP5, TP7, TP14 and TP15. This suggests that there was an east-west spread of material across the middle of the site. A sherd of Roman pottery was recovered from L4 in TP3. The burnt daub and Roman brick/tile probably derive from one or more tile kilns located on or near the site. A layer of post-Roman topsoil (L3) was observed in all the test-pits. This was approximately 0.25 m thick and was greyish-brown in colour, occasionally with an olive hue. In TP6, L3 was contaminated by modern petroleum spillages or leakages. The existing surfaces (L1) were of concrete and tarmac and were approximately 0.15m thick. They were usually laid on modern rubbly make-up (L2), 0.1-0.15m thick.