This research project was funded by the UK Economic and Social Research Council. The research was intended to investigate the belief that incoming undergraduate students of French in British universities have a declining knowledge about language, and in particular that their knowledge of metalinguistic terminology is very variable. The second main purpose of the research was to investigate the relationship between students' metalinguistic knowledge, and their language aptitude and language proficiency. A battery of tests was developed to study these matters, and administered to first-year students of French at seven UK universities In addition, Lancaster first year students were retested at the end of their first year, and second and fourth year students were also tested, to investigate any change in abilities or knowledge. The main findings are that students' knowledge of metalanguage is highly variable, and there are very few metalinguistic terms which students can confidently be assumed to 'know'. However, there is very little relationship between knowledge about language and language proficiency, at least for this population. This latter conclusion must lead to a questioning of the pedagogic value of teaching, using or focussing upon metalinguistic terminology, for improvement in language proficiency. For further information, contact Charles Alderson.