The evaluation consisted of a series of trenches spread across the site. These were mostly 50m and 100m in length, but also including one 80m trench. Some were positioned over anomalies and features revealed in geophysical survey; others to obtain a reasonable coverage of the evaluated area in apparently blank areas. Trench positions were laid out using a GPS. All trenches were initially dug by 360 machine with a toothless bucket under close, constant archaeological supervision. Topsoil and subsoils were excavated and stored separately, and were replaced in their appropriate stratigraphic order once investigation was completed. Trenches were generally clean enough to photograph and plan following machining; a few were subsequently cleaned by hand in part to clarify possible features. Empty trenches (in this case only Trench 5) were not planned. Following planning at 1:50 and photography, possible archaeological features were sample-excavated by hand to investigate their depth, form, character and date. Plans and sections were drawn of hand-excavated interventions on permatrace. All trenches were levelled along their length. An archaeological evaluation was carried out at Springfield Farm, Ambrosden between 16th and 28th January 2013. The area of proposed development currently consists of open grassland, and is currently being considered for development for housing. Following a desk-based assessment of archaeological potential, and geophysical survey, an archaeological evaluation was carried out prior to the submission of a planning application. The scope of the evaluation was agreed with Richard Oram, planning archaeologist for Oxfordshire County Council, and consisted of 12 trenches varying between 50m and 100m in length, and totalling a 2% sample of the proposed development area. Trenches were laid out to target anomalies identified by the geophysical survey, and to provide an even coverage of the development area. The work was carried out over 9 working days in adverse conditions with a team of 2-7 archaeologists and a 360� mechanical excavator. The evaluation revealed several ditches in the north-east corner of the site, two of which included single sherds of Roman pottery. One other Roman sherd was found in a post-medieval ditch. The paucity of Roman sherds, and the absence of any other material of this date, indicates that these features were probably field boundaries remote from settlement. These ditches were cut by the furrows of medieval or pos...