tT IS well known that the phonemic structure of English and Spanish X differs sharply in respect to the nasal sounds. English has three nasal sounds, [m], [n], and [w3], all of them phonemic entities, although the last of these is a defective phoneme since it does not occur initially. Otherwise minimal contrasts occur in all positions, witness meat. Eneat, bum bun bung, and simmer. Esinner singer. Spanish has the three nasal phonemes, Xm/, /n/, and /p/. Of these, the last is a palatal nasal, occurring most often in medial position, initially probably in not more than a dozen words in the whole language, and finally not at all. In fact, it need concern us no further. The phoneme /m/ is also defective in that it does not occur in final position. The velar nasal [4], although frequent enough in actual occurrence, is not a phoneme at all, but a regular positional variant of /nX before the velar stops [g] and [k]. The phoneme /nX remains, therefore, as the only one which is found initially, medially, and finally. We are faced then with the curious circumstance that although the phones [m], [n], and [q] are present in both English and Spanish, where they do not differ appreciably in the mechanics of articulation, yet they occupy totally different positions in respect to the phonemic structure of the two languages concerned. This raises the question: What effect does this difference in phonemic structure have upon the teaching problem, particularly that of teaching English as a foreign language to native speakers of Spanish? From the point of view of a spoken command of the language, that is to say of pronunciation, the chief problems are to teach the native Spanish speaker to pronounce [m] in final position, and to be able to make pq'1 in isolation without the accompanying velar stops. At first glance this might seem to be wholly a problem of oral production, but some of our experiences at the English Language Institute in Mexico in teaching English to students with a Spanish language background led us to believe that the hearing factor was also involved to a consideratble extent. It stands to reason that if the student does not hear accurately, imitation will not be an efficient teaching procedure, and even detailed phonetic instructions as to how to produce the sounds in question may fail of the desired effect. We set out, therefore, to devise a test of aural perception of