This paper explores the potential of using taphonomic analysis to reconstruct broad-scale variation in patterns of consumption and deposition at six later prehistoric midden sites in the UK. These sites comprise large accumulations of material culture, dominated by faunal and ceramic fragments, presumed to result from feasting events during the Bronze Age-Iron Age transition. New artefact and landscape studies have refined the characterisation of these sites (Tubb 2011a, b; Waddington 2009), but little research has focussed on accumulation history. This research uses simple statistical analyses on a large dataset (NISP >20,000) to compare the prevalence of bone modifications between midden sites. Crucially, significant differences in modification are not interpreted directly. Additional tests are undertaken to assess whether variation in assemblage composition could account for differences in modification. Previous research has demonstrated that certain elements and the remains of certain taxa are more likely to exhibit modification, and consequently, a prevalence of these specimens could account for differences, rather than their resulting from varied depositional treatment (Madgwick and Mulville 2012). Therefore, patterns of modification are only interpreted once compositional differences can be discounted from responsibility. The study is intentionally broad in its focus and assesses whether large-scale inter-site differences in depositional practice can be reconstructed. Clear patterns are observed with some middens accumulating predominantly through rapid, large-scale deposits and others building up through smaller, more gradual deposits and being subject to greater disturbance and bone movement. These findings have implications for our understanding of ritualised consumption and deposition at the Bronze Age-Iron Age transition in Britain.