The archaeological remains at Roughground Farm cover an area of c 8 hectares on the second gravel terrace just north of Lechlade between the rivers Leach and Thames (SP 216/009 to 221/005). The site was investigated by Margaret Jones in advance of gravel extraction between 1957 and 1965. These excavations revealed evidence of occupation from the Late Neolithic to the end of the Roman period and represent one of the first landscape studies undertaken in this country. The work was stimulated by the discovery of a Roman villa, whose buildings were partly investigated in 1957 and 1959. Further excavations on the villa buildings were carried out by Tim Allen in 1981-2 and in 1990 prior to a housing development.The Neolithic occupation consists of a small cluster of pits containing Grooved Ware, contrasting with a dispersed scatter of pits with Beaker pottery. The Earlier Bronze Age is only represented by a stray sherd, but there is a wide scatter of Later Bronze Age pits, which tend to congregate in small groups. In the Early Iron Age the landscape was divided by large boundary ditches, roughly parallel to one another and at right angles to the river Leach, with smaller ditched subdivisions. This land-division appears to respect established trackways, which met within the excavated area. Pit groups indicate an arable economy and occupation, including posthole groups and burials, was concentrated at the east edge of the site.The Middle and Late Iron Ages are hardly represented, but an Early Roman native settlement was established just west of the trackways. This included an oval house-enclosure with accompanying pit-group, small stock enclosures, and pens, lying within a larger rectilinear enclosure. Between the trackways and the settlement was an open 'green'-like area. The economy was similar to that of the Iron Age and this settlement persisted until the early 2nd century AD, when it was replaced by the building of a villa.At least two masonry buildings were put up in the mid 2nd century and were surrounded by an enclosure ditch. One of these was an aisled building, with^an apsidal end unique in Roman Britain. Outside this was a regular system of paddocks and larger fields laid out to a standard unit of length. The villa occupation area, however, kept within the limits of the preceding native settlement. Trackways and droveways approaching the villa were delineated by boundary ditches.In the 3rd century another large domestic building was constructed, while the ends of the trackways east of the villa were overlaid by two groups of enclosures facing each other across the 'green', which were used for various agricultural and semi-industrial activities and may also have been occupied. These may represent centralisation of the villa's estate management. Small groups of late Roman burials were found in and around these enclosures. In the 4th century, if not before, another domestic building was added to the villa. Occupation of the villa and adjacent enclosures continued beyond 360 AD, but possibly not as late as the end of the 4th century.There was very little evidence of Saxon activity, although the villa buildings were robbed for stone for graves in this period. The east part of the site was overlaid by ridge and furrow in the medieval period and the west appears to have been pasture; both parts remained open fields until gravel extraction began in the 1930s. Virtually the whole site has now been destroyed.