250 Mr. Willis on the Vowel Sounds, 



sion all the vowel qualities, as the eft'ective length of the spring 

 is altered, and that with considerable distinctness, when due 

 allowance is made for the harsh and disagreeable quality of the 

 sound itself. 



' But there is another remarkable phenomenon to be observed 

 in the experiment with Fig. 6, which, as it is wholly uncon- 

 nected with the vowels, I have hitherto, to prevent confusion, 

 omitted to notice. I stated that during the experiment the reed 

 retained its pitch, in fact however there is a considerable change 

 in it which I proceed to describe. 



In Fig. 18, the indefinite line aic6'... is taken, as before, to 

 represent the measuring rod; the ordinates of the curve, traced 

 below this line as an axis, are the lengths of the pitch-pipe in 

 unison with the note produced by the apparatus, when the ex- 

 tremity of the pipe reaches their respective feet*. Let ah, be, cb'... 

 be respectively equal to the half-wave of the reed, therefore c 

 is the center of the second series of vowels. 



Let the pipe be gradually drawn out; the pitch will remain 

 (jonstant till its extremity approaches f, about midway between 

 a and b. Here the note begins to flatten (which is represented 

 by the increase of the ordinates), proceeding still farther, it con- 

 tinues to flatten till the pipe reaches a point n beyond b, where 

 the note suddenly leaps back to one about a quarter tone sharper 

 than the original, which however it soon after drops gradually 

 to, and preserves till the pipe reaches c ; proceeding from c through 

 the successive cycles, the same phenomena will be found at the 



* To save room ou the plate, but at the same time to shew the real nature of this 

 change of pitch, these ordinates were constructed for the particular case of a reed (pitch 

 = 6.6) and drawn to the same scale as the rod (| of the original) but taken equal, not 

 to the length of the pitch-pipe, but to the excess of its length above six inches. If the 

 line ab were drawn an inch and a half higher up, the ordinates would be exactly pro- 

 portional to the real ones. (See Note A.) 



