Se ———eeeerl 
_ manipulating his piano-player. 
NoveEMBER 6, 1913] 
NATURE 
293 
or rather tension, acting on the ‘“‘ playing pneumatics ”’ 
or small bellows which operate the fingers, and the 
duration of the impulse necessary to produce the 
maximum effect is greater for the bass than the treble 
hammers. Now the regulating bellows are in a con- 
tinual state of vibration, producing rapid fluctuations 
of tension every time a note is played. If these 
fluctuations synchronise with the impulses required 
to produce the maximum effect in a particular part 
of the scale, it is evident that the corresponding part 
of the chord will predominate. Now, in playing 
softly the regulating bellows are only slightly com- 
pressed, and they open and shut slowly; in playing 
loudly they are much more highly compressed, and they 
collapse and open sharply. Thus the unpleasing want 
of balance in the quality of chords is easily accounted 
for. This difficulty I get over by varying the load 
on the regulating bellows, and also the inertia by 
means of a sliding weight, as well as by controlling 
its vibrations by hand. I have seen a patent in which 
it is sought to control the regulating bellows by en- 
closing it in an air chamber in which the tension can 
be varied by means of valves, and attempts have also 
been made to vary the tension in a spring controlling 
the bellows; but this can only be done by compressing 
the spring by a corresponding amount, and the time 
required to produce this compression appears to be 
too long to render the method efficacious. 
The usual method of concealing this want of 
balance is to operate the two halves of the keyboard 
with separate controls. This system produces effects 
which are pleasing at first, but are very arti- 
ficial and limited in character, and a person who is 
accustomed to this method of ‘‘faking"’ his chords is 
scarcely likely ever to learn how to balance their 
different parts with due regard to the effects required. 
Possibly this explanation may clear up some of the 
obscure points in my descriptions referred to by Prof. 
Morton. At the same time, I have heard 
professional pianists of considerable reputation whose 
range of control did not extend beyond that obtain- 
able by damping down the halves of the keyboard or 
accenting notes by means of punch-holes. 
Prof. Pickering’s references to the sustaining pedal 
or lever are calculated to suggest the inference that 
Prof. Pickering may not have had much experience in 
If he finds it necessary 
to listen for each note before he knows when to 
operate his sustaining lever it would appear that he 
has not yet learnt to play each note at the exact in- 
stant that he wishes it played, and in this case it is 
not easy to see how it would be possible to play 
accompaniments in which it is necessary to listen to 
and keep in time with the soloist. Personally, I 
have always considered that the sustaining lever 
played a far more important part in pneumatic playing 
than in hand playing, one reason being that the 
_ mecessary movements can be regulated with much 
greater rapidity and precision by hand than by foot. 
The right hand operating the speed regulator fixes 
the exact instant at which each note is going to be 
played, the left hand operates the sustaining levers 
and other controls at the correct predetermined in- 
stant. Probably, as Prof. Pickering says, an experi- 
enced pianist can also work hand and foot together, 
and I have known one musician who could operate 
the sustaining pedal of the piano three times in suc- 
cession in holding down a single chord. This would 
be quite easy with a piano-player, and I certainly 
_ the same in playing with fingers, 
_ often use the lever twice if not three times in playing 
a chord. But it must be much more difficult to do 
_ sa and with the 
majority of amateurs the main use of the loud pedal 
NO. 2297, VOL. 92] 
appears to be to compensate for the loss of resonance 
caused by boxing up the piano and covering it with 
rugs, vases, and photograph frames. 
The slight sound of suction through the air-holes is, 
of course, an inevitable defect, but one soon ceases 
to notice it. As regards ‘‘thud,’’ well, fingers as well 
as pneumatics sometimes produce this. 
G. H. Bryan. 
The Light Energy Required to Produce the Photographic 
Latent Image. 
THE amount of light energy required to produce a 
latent image on a modern high-speed photographic 
plate is known to be extremely small. The energy 
per silver grain may be roughly calculated without 
difficulty, and the calculation leads to some interest- 
ing conclusions regarding the nature of the latent 
image. 
Consider an exposure sufficient to produce a deposit 
of unit density, that is, one which will transmit but 
one-tenth of the incident light. A negative has unit 
density when the silver deposit is to milligrams per 
square decimetre, or o-1 mg. per sq. cm. (Sheppard 
and Mees, ‘Investigations of the Theory of the 
Photographic Process,” p. 41). This amount of silver 
represents roughly 10!? molecules, or 107 grains 3 » 
in diameter. Now the amount of light energy re- 
quired to produce an exposure giving unit density is 
of the order of 107 watt-sec. (erg) per sq. cm., and 
therefore 10-™ erg per grain, or 10-*° erg per mole- 
cule. The probable uncertainty in these values is not 
greater than a factor of 10. 
The effect of the ligkt on the plate is to permit 
the chemical reduction of silver halide to metallic 
silver with an additional expenditure of energy less 
than that required to reduce the unexposed silver 
bromide. Development we know to proceed by whole- 
grain units, hence we reason that one molecule in a 
grain (10! molecules) is so affected by exposure that 
the whole grain is developable. 
The simplest assumption to be made is that one 
electron per grain is detached from one molecule; 
such a liberation would require (Davis, Phys. Rev., 
XX., Pp. 145, and others) 5 x 10-1 erg, or less (Astroph. 
Journ., Xxi., p. 404), a quantity consistent with that 
calculated above from the known exposure and mass 
of silver. Hence the hypothesis is reasonable that the 
latent image consists of halide salt in each grain 
of which one electron has been liberated by exposure 
to light. P. G. Nurtinc. 
Research Laboratory, Eastman Kodak Co., 
Rochester. 
An Aural Illusion. 
Mr. ALLISTON refers in NaTuRE of September 18 
(p. 61) to a certain aural illusion, and wonders if 
anyone has thought of it before. 
Two or three years ago, in a letter to Knowledge, I 
commented on the fact that if a flash of lightning 
2 or 3 miles long happened to occur ‘‘head on” to an 
observer, the result of the flash travelling so far 
quicker than the sound would be that he would hear 
first the thunder caused by the part of the flash 
nearest to him, which arose last, and then in succes- 
sion the earlier sounds, until finally he would hear 
the opening crash, like a phonographic record 
reversed. Sometimes I have noticed that a thunder 
peal ends up with a sudden and more violent crash, 
and I wonder if this is owing to the explosion which 
begins a peal of thunder being louder and more 
abrupt than the after noise T. B. BratHwayrt. 
Cape Town, October to. 
