1Technical Aspects of Visible Speech, in Bell Telephone System Technical Publications, Monograph B-1415, pp. 19-49. Reprinted from the Journal of the Acoustical Society of America, xvII (July 1946), 1-89. 2 Language, Vol. 24, No. 2 (April-June 1948), Supplement. Language Monograph No. 23. 3R. K. Potter, G. A. Kopp, and Harriet C. Green, Visible Speech. New York: Van Nostrand, 1947. 441 pp. 4 In Technical Aspects of Visible Speech. See above, Note 1. Pages 62-73. 5 reader will find detailed experimental proof of this theory in Gordon Peterson's article, The Phonetic Value of Vowels, in Language, Vol. 27, No. 4 (Oct.-Dec. 1951), 541-543. 6 Assuming the maximum possible syllabic flexibility, we have (16+16+162) variations: 288. 7 See Schultee, F. C. and Lorrine Torrez: Two Rules in Need of Revision (x before a consona t as [s]), Hispania, xxx (1947), 209; Richard L. Predmore, The Pronunciation of x Before Another Consonant, Hispania, xxxI (1948), 196-197; Dwight L. Bolinger, That x Again, Hispania, xxxi (1948), 449-450; R. L. Predmore, One More Look at the Pronunciation of x Before a Consonant, Hispania, xxxII (1949), 344-345; and D. Lincoln Canfield, Sinfonia en s menor, Hispania, xxxIv (1951), 288. 8 For prices and terms, write to the Kay Electric Co., Pine Brook, N.J., manufacturers of precision electronic equipment. 9 By means of a rather complicated mathematical treatment of the data, it is possible to adjust for the personal idiosyncrasies of the spectrogram. This permits direct comparison of data collected from different speakers of the same language. method is outlined in the following articles: Gordon E. Peterson, Vocal Gestures, in the Bell Telephone Laboratories Record (Nov. 1951), 500-504; by the same author, The Phonetic Value of Vowels, in Language, Vol. 27, No. 4 (Oct.-Dec. 1951), 541-553; Gordon E. Peterson and Harold L. Barney, Control Methods Used in a Study of the Vowels, in Journal of the Acoustical Society of America, Vol. 24, No. 2 (March 1952), 175-184. In a study now in preparation, the author of this article plans to present such a de-personalized analysis of a complete set of data in Spanish recorded by Professor Toms Navarro.