RESEARCH ARTICLE A refitting experiment on long bone identification Eugène Morin 1,2 | Arianne Boileau 3 | Elspeth Ready 4 1 Department of Anthropology, Trent University, Peterborough, Ontario, Canada 2 PACEA, Université de Bordeaux, Talence, France 3 Department of Anthropology, University of Florida, Gainesville, Florida, USA 4 Department of Human Behavior, Ecology and Culture, Max Planck Institute for Evolutionary Anthropology, Leipzig, Germany Correspondence Eugène Morin, Department of Anthropology, Trent University, DNA Bldg Block C, 2140 East Bank Drive, Peterborough, Ontario K9J 7B8, Canada. Email: eugenemorin@trentu.ca Funding information Social Sciences and Humanities Research Council of Canada, Grant/Award Number: #435-2013-0993 Abstract Refitting is an important analytical tool in archaeology that can yield valuable infor- mation on site formation processes and on the range of activities practiced at a site, including tool production, tool curation, and discard behavior, among others. In the present paper, we use refit data from a control assemblage of red deer (Cervus elaphus) long bones to assess problems of specimen identification and repre- sentation in an experiment where bones were processed for marrow. Three goals motivated this experiment: (i) to assess how different methods of NISP (number of identified specimens) calculation affect comparisons of the relative abundances of long bone regions, (ii) to evaluate whether long bone shaft regions vary with respect to the probability of identification, and (iii) to ascertain the potential refit rate for a well-preserved and fully-collected sample of faunal specimens. Our results show no statistical differences in terms of patterns of skeletal representation between the two methods of NISP calculation (single vs. multiple NISP counts) that we assessed. Our data also indicate that the shape, particularly the cross-section, of fragments clearly impacts the probability of identification and refitting. Moreover, the refitting experiment reveals that, in ideal conditions, a majority of specimens (>95%) from the NISP sample can be refitted, which leads to largely reconstructed skeletal elements. Thus, the comparatively very low refit rates recorded in archaeological sites, including samples that are well preserved, suggest that the often limited extent of excavations, along with offsite discard and/or extensive sharing of parts, substantially reduce the possibility of finding refits in a faunal sample. KEYWORDS faunal identification, faunal quantification, refitting 1 | INTRODUCTION Identifying faunal remains to skeletal element, and where possible, to taxon plays a pivotal role in most archaeozoological interpretations (Gifford-Gonzalez, 2018; Nims & Butler, 2017; Reitz & Wing, 2008). However, substantial evidence suggests that certain categories of skeletal parts are easier to identify than others, a finding that affects both the accuracy and reproducibility of archaeozoological inferences (Hudson, 1990; Morin et al., 2017a, 2017b; Pickering et al., 2006). Potential biases in interpretation due to differential identifiability seem particularly acute in contexts where the articular ends of long bones have been deleted by carnivores or other agents. This is because, in this context, identifications must be derived from the shaft portions of the bones, which are allegedly more difficult to identify (Bunn, 1991; Costamagno et al., 2005; Marean & Kim, 1998; Morin, 2010; Pickering et al., 2006; Turner, 1989). Several studies suggest that extensive refitting can alleviate problems of under- representation of long bones and long bone shafts in these contexts, although whether this can substantially modify patterns of skeletal representation remains controversial (Bartram & Marean, 1999; Klein et al., 1999; Marean & Frey, 1997; Marean & Kim, 1998; Pickering et al., 2003; Stiner, 2002; Yravedra & Domínguez-Rodrigo, 2009). Received: 8 January 2021 Revised: 5 March 2021 Accepted: 9 March 2021 DOI: 10.1002/oa.2980 Int J Osteoarchaeol. 2021;1–13. wileyonlinelibrary.com/journal/oa © 2021 John Wiley & Sons, Ltd. 1