In this paper, we will conduct a preliminary investigation into a relatively new approach to terminography: corpus-based terminography. This subject has thus far received very little attention in the terminology literature, so we will begin by briefly investigating the evolution of corpus-based research in the fields of linguistics and lexicography. We will then move on to examine in more detail the development of the emerging field of corpus-based terminography, including the advantages of adopting a corpus-based approach and the requirements for compiling machine-readable corpora that are useful for termino graphical purposes. Finally, we will briefly consider the tools and methodology required for retrieving terminologically interesting information from such corpora.