January 6, 1923] 
NATURE 21 
a awe _ mia ozr si 
of the surface of the earth which determines our 
standard time. I find it difficult to believe that there 
can be irregular variations in the angular velocity of 
the earth as a whole ; but it seems less difficult if the 
variations are merely superficial, due to the crust 
sliding non-uniformly on the interior. I have even 
entertained the wild idea that the motion of the 
magnetic poles might be due to this cause ; the mag- 
netism being constant in the interior but with the axis 
emerging at changing points of the crust as the crust 
slips over the inner magnet. Unfortunately, so little 
seems to be known about the motion of the magnetic 
poles that I have not even been able to make out 
whether the motion is from west to east as this theory 
definitely requires. 
What interests the geologist more nearly is that the 
brake is applied only at certain areas on the surface, 
so that there would be a tendency to crumple the crust 
more particularly to the west of these areas. It is 
unfortunate that shallow seas are necessarily the least 
permanent features of the earth ; otherwise I would 
have asked whether the geologists had evidence of 
special crumpling in such areas. 
I have regarded the crust as fairly mobile from east 
to west. I suppose the geologists would also like it 
mobile from north to south in order to have glacial 
periods in those portions which are now near the 
equator. It is not possible to hold out much en- 
couragement for such an idea, because we cannot 
imagine any force acting from north to south. Still 
if the crust, which is being urged by the east-west 
force of tidal friction, is resisted by obstacles it may be 
deflected, finding that say a south-west track offers 
less resistance. In a long enough time almost any 
displacement may have happened, granting my 
hypothesis that the connexion of the crust to the 
interior is reasonably plastic. So I cannot forbid this 
possible interpretation of glacial periods in the earlier 
geological times. 
Iam sure that it will not be supposed that, in present- 
ing the astronomical side of these questions which belong 
both to geology and astronomy, I have any intention 
of laying down the law. The time has gone by when 
the physicist prescribed dictatorially what theories 
the geologist might be permitted to consider. You 
have your own clues to follow out to elucidate these 
problems, and your clues may be better than ours for 
leading towards the truth. We both recognise that 
we are adventuring in regions of extreme uncertainty 
where future discoveries will probably lead to various 
modifications of ideas. Where, as in the new views of 
the age of the earth, physics, biology, geology, astro- 
nomy, all seem to be leading in the same direction, and 
producing evidence for a greatly extended time-scale, 
we may feel more confidence that a permanent advance 
is being made. Where our clues seem to be opposed, 
it is not for one of us to dictate to the other, but to 
accept with thankfulness the warning from a neighbour- 
ing science that all may not be so certain and straight- 
forward as our own one-sided view seemed to indicate. 
Nature and Reproduction of Speech Sounds.! 
By Sir RicHarp Pacet, Bart. 
AM the characteristics of English speech — the 
vowels and diphthongs and consonant sounds— 
can be produced—as breathed or whispered speech— 
without using the larynx at all; so that in the use 
of the English language (at least) it may be said that the 
larynx is not an essential organ of speech. The function 
of the larynx is to give carrying power and inflexion 
to speech, and melody to song—it has nothing to do 
with the essential characteristics of speech. 
If any one with a normal “ear for music” will 
whisper the words “ Noah’s rather at sea ’—thinking 
of the sounds rather than of the sense—they will hear 
BERS 5 On ae 
oS Qa @-2* t 
Noah's rather at Sea. 
Fic. 1. 
an ascending arpeggio something like the phrase shown 
in Fig. 1. The exact notes heard in each case will 
depend on how the individual person pronounces the 
vowel sounds in question. 
These whispered or breathed notes are formed, as 
is well known, by the resonance of the cavity of the 
mouth, and they are varied for each different vowel 
1 Substance of a lecture delivered at University College, University of_ 
on October 18. 
NO. 2775, VOL. 111} 
sound by altering the size of the cavity and the opening 
of the mouth, mainly through the operation of the 
tongue and lips. With many of the vowel sounds, 
namely, i (eat), ei (hay), e (men), # (hat), o (not), and 
in some types of a (calm), two simultaneous resonant 
notes have been heard by many investigators, but the 
remaining principal vowel sounds, 9 (all), ou (no), and 
u (who), have been generally supposed to be character- 
ised by a single resonance. 
Some observations made by me at the beginning of 
this year, using my own breathed vowel sounds, 
indicated that in every case the mouth—or rather the 
oral cavity as a whole, from the larynx to the lips— 
actually gives two simultaneous resonances for each 
vowel sound. It appeared that these pairs of resonant 
notes are not fixed in pitch for any one vowel sound, 
but might vary over three or four semitones—and some- 
times even more—without a very appreciable change 
in the character of the vowel. 
The resonances heard in the use of my own voice 
are set out in the accompanying chart, in which the 
vertical scale represents semitones of the equal tempera- 
ment scale, and the vowel sounds are represented in 
the notation of the International Phonetic Association 
(Fig. 2). 
It will be seen that i (eat), I (it), ei (hay), e (men), 
z (hat), @ (earth), 3 (sofa), a (up), and a (calm) form 
very nearly a converging series—the upper resonances 
falling by steps of 1 to 3 semitones, while the lower 
resonances are more active and take larger jumps— 
