774 PROFESSOR FLEEMING JENKIN AND J. A. EWING ON THE 
difference between vowel sounds uttered in speech and in song may conceivably 
be that in the former the cavity is constant, and in the latter it is tuned. 
There is some evidence in the analyses that even the 6 cavity is not quite 
constant. We have already drawn attention to the apparent abruptness with 
which some of the constituent partials come into, or disappear from, prominence, 
and this might be explained by supposing a certain adjustment or tuning of the 
cavity. The analyses in Table V. of 6's, sung by voice 6, appear to indicate 
that, in examples 53 and 54, a strong upper proper tone of the cavity was in 
unison, or very nearly in unison, with the fourth partial of each. In example 
55 it would appear from the proportion which the amplitude of the fourth 
partial bears to that of the fifth, that this proper tone was still much nearer to 
the fourth than the fifth partial; it seems to be following the fourth partial 
down the scale. But in example 56 it has apparently risen so as to be exactly, 
or very approximately, in unison with the fifth partial. 
If we assume that the 6 cavity is absolutely constant, we must describe it as 
a cavity reinforcing tones throughout nearly two octaves, or from g to /”, for 
we have found in some cases a strong second partial when the prime is on G 
and on /’. Weare disposed to regard it as more probable, that in human 
voices the 6 cavity is slightly tuned or modified according to the pitch on which 
the vowel is sung. We do not mean by this that the cavity adapts itself so as 
to be in unison with some particular partial, but only that when no partial falls 
on b,’, the cavity will make its pitch of maximum resonance deviate a little 
from that pitch, so as to approach more nearly, and therefore reinforce more 
strongly, some partial tone lying near. And the same thing may occur with 
other proper tones of the cavity than the tone b,. If we regard the 6 cavity 
as adjustable in this manner, we should describe it as a cavity capable of rein- 
forcing tones over somewhat more than one octave. The range of adjustment 
need not exceed about six semitones, and the upper resonance would never 
deviate far from b,’, on which pitch the strongest reinforcement can be given. 
Another way of putting the same view would be to say that the genuine char- 
acter of 6 is given by a cavity reinforcing tones over rather more than one 
octave, with an upper proper tone never far from b,, and that the human voice 
in singing may choose (to suit the pitch) an 6 cavity, whose upper proper tone 
deviates slightly from b,’, in order to bring some one strong proper tone of the 
cavity more nearly into unison with one of the partial tones of the note sung. 
Of course, this might be explained by saying that the voice chooses a different 
species of the generic vowel 6 at different notes, so as to bring out as loud a 
resonance as possible. It is to be observed that the hypothesis of a tuned 
cavity is much strengthened by the evidence which exists for believing it to be 
tuned in the vowel i, evidence which, it must be admitted, is much stronger 
than any which we have obtained for 6. 

7 
7 
ee ee ee ee ae 
wr 
> =v 
