One of the most popular sub-disciplines of archaeology is experimental archaeology, the re-creation of items, structures, and practices of past societies in the present day. This area of study has a long pedigree in Continental Europe, but was a relatively late development in Britain. One of the pioneers of this approach was Peter Reynolds, who created the Butser Ancient Farm Research Project to explore life in the Iron Age (Reynolds 1979). When it was set up, in the 1970s, experimental archaeology was undertaken with full scientific rigour. Important goals included the quantification of resources required to create a house, the management of ancient breeds of domestic animals, the productivity of Welds of ancient cereals, and the function of pits. All these tasks were carried out with a critical attention to detailed data recording and scientific rigour. More recently, experimental archaeology has become geared towards the general public, and though Butser Farm has retained a scientific core to its activities it also caters for a wider public, providing both knowledge and entertainment about past societies. I had a brief experience of this work in 1977 when I took part in a week-long Weld school at Butser Farm, organized by Glasgow University. This was a key period in the development of the Iron Age farm. The original farm had been created on a spur near the top of Butser, specifically away from easy public access and in a very exposed location. Public interest in the experiment had become difficult to manage and a new site had just been located in the Queen Elizabeth Country Park, a much more accessible location near the main road from Portsmouth to London. The new location was designed to be a public amenity that would attract visitors to the Country Park and represented a move from ‘Laboratory to Living Museum’ (Reynolds 1979: 93). The main job we were to undertake was to help with the construction of a large roundhouse that would form the centre of the new farm. Two previous timber houses had been built up on the hill, but both had been fairly modest affairs; one was based on a house plan from Wheeler’s excavation at Maiden Castle, the other was slightly larger and based on the excavation of a house in the Balksbury enclosure, Hampshire (Reynolds 1979).