This chapter describes a new longitudinal corpus of L2 Spanish, which includes spoken and written data from the same learners, and was gathered in a residence/ study abroad setting. The corpus was created as part of the Languages and Social Networks Abroad project (LANGSNAP) which traced the learning of both French and Spanish during an academic year's residence/ study abroad by British Anglophone university students majoring in languages. Parallel longitudinal learner corpora totalling 300,000+ words were collected in each language. In this chapter we first describe the main features of the L2 Spanish corpus and then illustrate how we have been using the corpus to investigate questions about language development, taking as an example the longitudinal development of lexical diversity (Jarvis 2013).