This chapter looks at how computer-based corpora – collections of translated text - can be used to study trends in translation practice. The aim is to outline what can be regarded as some of the main current and potential uses of corpora in the empirical study of translation, translator training and professional translating.