This paper reports research into the acquisition of vocabulary in a foreign language, more specifically into the supposition that 'guessing' (i.e. inferring the meaning of an unknown word from context and word-form) contributes substantially to retention. Although much is to be said for the latter, the empirical underpinning of this supposition has so far been slight and various questions have remained. In order to gain a better insight into these we have carried out an experiment in a classroom setting, focussing on the following questions:1.Which factors of context influence the guessability of words?2.What is the influence of these factors on receptive retention (after guessing and learning)?3.What is the relationship between guessing and retention? (Are words which have been correctly guessed better retained than words which have not been correctly guessed?)The chief findings of the experiment are:l.A specific ('pregnant') filler of the slots 'subject', 'verb' and 'function' contributes to the guessability of a word in a particular sentential context.2. A specific ('pregnant') filler of the slots mentioned in guessing and learning has no effect on the retention ('subject' and 'verb') or even a negative effect ('function').3.There is no positive correlation between guessing and retention (after guessing and learning; for some words there is even a negative correlation.To sum up: a 'pregnant' context does indeed induce better guessing, but subsequently it does not induce a better retention of the meaning involved, at least not if the guessing stage is followed by a learning stage using the same context as in guessing.