224 Proceedings of Royal Society of Edinburgh. [sess. 
is felt very accurately. When the resonance of / is carefully 
listened to, it is found to be far from constant in pitch. The 
resonance of / in fee has, in my own case, the pitch a, octave 
uncertain ; that of / in fcdl is about / — five semitones lower. But 
it is best to study the consonant / apart from any vowel. 
It is a further good reason for beginning with / and those frica- 
tives which most resemble it, that they can be produced in isola- 
tion, and prolonged ad libitum without at all losing their charac- 
teristic effects on the ear. It is a long time before the phonetic 
observer acquires the conscious separate command of each of his 
organs — velum, nares, larynx and tongue. After the consciousness 
of their separate action comes the consciousness of their several 
positions, at any given time. This consciousness grows in accuracy 
by actual inspection and handling, and by nothing else. But it is 
at first quite unreliable, and must be trusted no further than it can 
be tested by the mirror, the probe or the finger, or by diagrams set 
out from the testimony of these. When these cautions are observed, 
the following observations may be carried out concerning /. 
When freed from connection with any vowel, the resonance of / 
can be carried a long way both up and down in pitch, without at 
all spoiling the / itself. Its whole range in the individual speaker 
is about two octaves — an octave each way from the / in fall. Add 
to this range another octave for the difference between infant and 
adult resonances, and it becomes clear that the essential quality of 
/ is but vaguely linked with the actual pitch of its resonance. The 
quest made in that direction is therefore a vain one : we must look 
elsewhere for the acoustic essence of /, or of any other fricative. 
But it is not vain to ask whence that resonance comes. It seems 
clearly to arise in that free passage between the tongue and palate 
which is an essential part of the articulation of /. This is evidenced 
by the fact that in the upper octave of its range this passage 
lengthens or shortens itself in proportion to the fall or rise of the 
pitch of the resonance. It is impossible to lengthen the passage 
any more than this ; and the further fall is secured by bulging the 
passage, till at last it becomes a considerable cavity. The upper 
octave, with its simple lengthening and shortening of tube, affords 
some criterion of the real pitch of the resonance. In my case, the 
tube at its longest is about 100 mm., at its shortest about 50 mm. 
