In an earlier study, Service (1992) found that phonological short-term memory, reflected in the ability to repeat English-sounding pseudowords, was a good predictor of learning English as a foreign language over a period of three school years. This study attempts to find out to what extent foreign vocabulary learning is the critical learning process related to pseudoword repetition accuracy. Regression analyses on repetition data and learning measures revealed significant correlations between pseudoword repetition and foreign language learning, even after a measure of general academic achievement had been partialed out. In a fixed-order multiple regression analysis – with predictors entered in the order of general academic achievement, foreign language vocabulary, and pseudoword repetition – no additional variance in any foreign language measure studied could be accounted for by pseudoword repetition in the third step. When vocabulary in the second step was replaced by performance in foreign language essay writing or the correct reproduction of dictated discourse – both highly related to overall foreign language performance – repetition significantly increased the proportion of accounted variance, especially for two foreign language vocabulary tasks and a listening comprehension task. Phonological memory may, therefore, be specifically related to foreign vocabulary learning.