This paper analyses the topic of representation since the point of view of ethnomethodology and sociology of scientific knowledge. It starts out by discussing the “standard image of representation” and the constructivist proposition of that image. Then, a case of study is presented to suggest how practices for collecting and analyzing forensic evidence in criminal law, can contribute to understand representational adequacy. The aim of this paper is to think differently about representantion considering how it is produced, managed and deconstructed. Since this point of view, the propositions is to consider representation not only as a problematic epistemological concept but also as a practical accomplishment in specific epistemic cultures.