The efficacy of teaching the pronunciation of English polysyllables to non-native speakers by means of ordered rules which use standard orthographic representations as underlying forms was tested on two groups consisting mainly of francophones. The experimental group applied ordered quasi-phonological rules to selected words from each of fifteen word classes being tested. The control group performed repetition exercises on these same words. The absolute improvement and relative improvement of the experimental group as compared to the control group were significant at the .02 and .002 level, respectively. Pedagogical and psycholinguistic implications are noted. (Author/CFM)