Résumé
Soil invertebrates are assumed to play a major role in ecosystem dynamics, since they con-tribute to ecosystem services. Despite their acknowledged importance, relationships betweensoil invertebrate diversity, soil processes and environmental changes deserve more attentionand still cannot be satisfactorily predicted. Taxonomic or a priori functional groups ap-proaches were usually used. Consequently, a new theoretical framework is proposed relyingon the concept of functional traits. Functional trait is de ned as any morphological, physi-ological, phenological or behavioural feature measurable at the individual level.Functional trait approaches have been intensively developed in several domains (e.g. plantecology). However, soil ecology stayed the poor relation of this drive. Bene ting from ex-perience in other domains, we have identi ed determinant obstacles to the development offunctional trait-based approaches for soil fauna.Consequently, we built two interoperable tools: a database (http://betsi.cesab.org/) and athesaurus of trait names (http://t-sita.cesab.org/BETSIvizIndex.jsp). Advantages of thissemantic data integration will be presented in terms of scienti c and methodological aspects.Future challenges will be addressed. Among them, the collective e ort that is pivotal tomove forward in such a framework. We can also cite the need for the construction of datawarehouse to bring together several trait databases or the interfacing of trait and barcodedatabases.