Phonozoic Text Archive, Document 106
Helmholtz's Vowel Theory and the Phonograph
Nature, Vol XVIII,
May 23, 1878, p. 93.
THE results obtained by Messrs. Jenkin and Ewing in their experiments with the phonograph, as described in NATURE, vol. xvii, p. 384, are so different from those reached in some experiments recently performed by Dr. Clarence J. Blake, of this city, in connection with myself, that I venture to call attention to the fact.
With the design of testing the question of change of quality in vowel tones by increasing the rate of rotation of the phonograph cylinder, we performed a number of experiments, of which I mention a few as briefly as possible.
1. The vowels ou and ó were spoken into the mouth-piece of the instrument, each four times in succession, while the cylinder was rotated at the rate of one revolution per second, as timed by the beats of a clock-pendulum. On rotating the disc so as to reproduce the vowel-sounds, these were spoken, ou, o, each repeated four times, when the rate of rotation was one revolution per second, but on increasing the velocity to two revolutions per second, the first sounds were indistinct, while the last gave the  vowel è [should be short!] very clearly. At half revolution per second, ou, au, were distinctly heard.
2. The vowel ó was sung while the cylinder was rotated at different rates of speed. On reproducing the sounds, the cylinder being revolved more slowly than at first, the vowel au was heard, changing to ó, ë, è, falling to è again as the velocity was slackened a little.
3. The vowel ä was spoken while the cylinder made one revolution per second. On reproducing the sound, the rate being half a revolution per second, au was heard, changing to ä when the rate increased to one revolution, and at three revolutions per second ì was heard.
4. The vowel ó was spoken several times in succession, the rate of the cylinder being gradually accelerated. On reproducing the sound by a uniform and slow rotation, au and ou were heard; on rotating faster, è and ì.
Several other experiments were tried in the short time during which the instrument was at our service, all of which were strikingly confirmatory of Helmholtz’s theory. Difficulty was experienced in reproducing the highest vowels é, ì, probably on account of want of readiness of response in the disc. The bell of a reed-pipe was placed over the mouth-piece of the instrument when the sound was to be reproduced, for which a horn of pasteboard was substituted in some of the trials.
We hope to render these experiments more rigorously quantitative, as the phonograph promises to be a valuable aid to research in this field. Very probably others may have worked with the same end in view, and if so it would be interesting to learn what has been their experience.
CHAS. R. CROSS
Boston, U.S., April 29