The substantial Yarnton report is one of three volumes (Vol. 1, Saxon and medieval material, appeared in 2004, while Vol. 3, Neolithic and Bronze Age, is still awaited) discussing a long and complex series of mainly large-scale excavations undertaken by Oxford Archaeology within an area of gravel extraction of some 140 ha, with some outliers, situated on the floodplain and the Second Terrace in the Upper Thames Valley. The area's archaeological potential had been known since the nineteenth century, from finds during railway and road construction, and earlier gravel digging, while in the twentieth century the cropmarks of complex, extensive and clearly multi-phase fields, enclosures and other features were recorded in the area by Major Allen in the 1930s and Professor St Joseph in the 1960s and 70s. This is thus a very complicated report on a whole series of separate interventions over a very large area.
Not all the 140 ha was archaeologically investigated, but the extent of some of the excavations has permitted the examination of the landscape on a much larger scale than for example the excavations at Farmoor or Port Meadow. We see here too the recognition of episodes of alluviation from flooding, followed by the regular re-establishment of the ditched field-systems. It is suggested that the Yarnton area was inhabited by ‘only two or three families’, whose shifting and intensifying patterns of exploitation are discussed in detail, with a move from open to enclosed settlement in the late Iron Age, as at Gravelly Guy. The use of the floodplain for grazing in the Middle Iron Age occasioned the construction of several very fine causeways of limestone laid upon an earlier wooden phase. In the early Roman period, while continuous land-use is clearly suggested, the landscape was substantially reorganised, although no early Roman structures appear to have survived. Land-use intensified and those parts of the floodplain less subject to flooding were ploughed for cereal cultivation. By the late Roman period the field patterns were further regularised, and buildings, corn-driers and so on were excavated.
It is impossible here to summarise adequately such an enormous report of various complex excavations, with reports on both Iron Age and later Roman cemeteries (the latter including cremations and inhumations) and the admirable palaeoenvironmental studies associated with Oxford Archaeology's work on the Thames gravels. Two early Roman pottery kilns were excavated, while the total Roman ceramic assemblage (discussed by Booth) comprised over 10,000 sherds. There was little in the way of imported fine wares or amphora, and few fine wares even of more local provenance. The material is mainly of local origin and the overall impression is of a small and relatively self-contained society throughout the Roman period.
The volume, and its companions, provides a splendid addition to the existing series of reports on the archaeology of this area. This is of great importance, as the whole history of the occupation of this part of the Upper Thames floodplain and gravels is fast vanishing into the trucks of the aggregates industry, and reports such as these will be the only basis for future study of much of the area. In this respect, it is notable that the planning permission for gravel extraction at Yarnton was granted prior to the introduction of Planning Policy Guidance 16 (Archaeology) in 1990, and hence no archaeological conditions were imposed upon the programme of gravel extraction. Although the site operators ARC (subsequently Hanson) provided help in kind in the form of plant etc., the bulk of this large and important programme of excavation and post-excavation was funded by English Heritage.
The second volume by Wessex Archaeology covers further investigations ahead of the expanding development of Poundbury (Dorset), a new ‘model’ town of the Duchy of Cornwall on chalkland lying between the Dorchester Roman aqueduct, Maiden Castle and the county town of Dorchester, where development and expansion has been more or less continuous since 1993 and is due to continue until 2025. This volume reports on the fieldwork undertaken at Poundbury Farm, a large-scale archaeological evaluation in 2006 followed by targeted fieldwork in 2007, and the results of 2005 excavation at neighbouring Poundbury Parkway.
The report follows a standard format, setting out the background to the work, the aim and objectives, and outline methodology. Presentation of evidence is divided between early prehistoric to Iron Age and Romano-British to post-Roman periods. A large chapter on finds includes specialist reports on coins, metalwork, flint and chert, worked stone, the stone coffin, Neolithic pottery, Romano-British pottery, post-Roman pottery, and other finds (including glass, shale, fired clay, CBM, slag and worked bone). There are also chapters on the human remains and the environmental evidence.
The fieldwork provided evidence of a multi-period landscape with activity concentrating in the Bronze Age and Romano-British periods, but with slight evidence of pre-Neolithic activity in the form of a Palaeolithic hand axe and Mesolithic flints (none found in situ), Neolithic settlement and agriculture, and post-Roman use. It is impossible here to summarise the detail for all these periods, but it is worth noting that the excavation recovered no Iron Age material, which is said to be the normal situation in this area. This is slightly surprising given the proximity of this site to Maiden Castle, Poundbury hillfort, and numerous Iron Age settlements whose relationship to Maiden Castle was discussed by N. Sharples some time ago. Certainly middle and late Iron Age evidence is widespread through much of central and southern Dorset. If Iron Age activity is indeed absent from this site, it is perhaps more unusual than herein suggested and could be a subject for further consideration.
There is a large amount of Romano-British evidence from this site, including small groups of Durotrigian burials, mainly from the early part of that period, although with some evidence for continuity of plot use and burial rite through the middle and late Romano-British periods. One burial was contained within a stone coffin that had been excavated in the late eighteenth/early nineteenth century then reburied. Another unusual example is of a cremation burial in an urn complete with its clay sealing plug. This permitted exceptionally well-preserved human remains to be recovered and assessed by McKinley. Closing activity at the site included the deposit of a small late Romano-British coin hoard. Thereafter the farmstead was abandoned and the area used for agriculture with a single sherd of Saxon pottery being important evidence for a period little represented in this area.
Overall this is a straightforward report on the results of fieldwork at Poundbury. Unusual features are presented, but their significance, and perhaps the lack of Iron Age evidence, is somewhat underplayed.
These two publications report on excavations of undoubted interest to readers of this journal. Yarnton has a much more detailed and complex history and also manages to draw more from the excavations than the report on Poundbury. Both provide useful catalogues of material and discussions of chalk and gravel landscapes to put their sites into the contexts of current prehistoric and Romano-British understanding.
