The data presented in this paper on the pronunciation of the syllable structure in English by Spanish speakers corroborate one of the main issues raised against the original formulation of the Contrastive Analysis Hypothesis: consonant sounds which seem to be similar in two phonetic systems can in fact be more difficult to acquire than sounds and sequences of sounds which are different. In other words, interference from NL to TL can actually be greater when items to be learned are more similar to existing items. In light of this deficiency, a revised version of the original CAH has been suggested that defines the subset of errors which are caused by transfer and which would predictjust what sorts of native language-target language differences will cause language learners to make errors. A possible step towards a revision of the CAH requires an investigation of the sorts of errors which can be shown to result from native language interference. Such an investigation should include a systematic exploration of the error in question, that is, not only detecüng the errors caused by native language transfer but also understanding the fundamental principles of the phonological processes involved both in the NL and TL.