454 PHOTOELECTRICITY, ART AND POLITICS 
is its speed. This advantage has been realised to some extent; but the 
extent has been greatly exaggerated. When photoelectricity is mentioned 
in the popular press one of the things that always turns up is a machine 
for sorting cigars and coffee beans according to their colour. That is, of 
course, very high speed measurement. Now sorting machines of this kind 
can undoubtedly be made; a beautiful example made by Mr. Horsfield 
was shown a little time ago in the Exhibition at the Science Museum. But 
they are emphatically not yet general; we have never yet been able to 
hear of one in regular use. The difficulty is that the qualities that should 
determine sorting in these cases, although simple enough to the eye, are 
intricate combinations of form and colour that mislead any less complex 
instrument. Developments in this direction are quite probable; but at 
present it is yet another direction in which enthusiasm has outrun discretion. 
The speed of action characteristic of photoelectric cells can be utilised 
only when little else is demanded. Use of it-is made, of course, in sound 
films ; for the cell has then to follow light vibrating with the frequency of 
sound, that is to say, thousands of times a second. Even greater demands 
on it are made in television. Perhaps you have been expecting us to say 
much on this subject, which is so topical ; but it is not of great interest from 
our present standpoint ; its implications, so far as they can be foreseen, 
have long been obvious to all, and are not very different from those of 
picture telegraphy, which has actually been achieved. The idea of tele- 
vision was in the minds of the earliest inventors ; and, as I have said, they 
formulated its problems quite correctly. For the last fifteen years at least 
all the fundamental problems have been solved in principle ; it has been 
clear that the attainment of television of almost any desired degree of 
excellence has been simply a question of expense. We must not appear to 
belittle the work of those who have achieved so much in this field ; if an 
engineer is one who can do for a shilling what any fool could do for a pound, 
then they have truly proved themselves engineers. Perhaps the most 
useful remark we can make is this ‘The long period of delay, while it has 
been doubtful whether the public would be attracted by such television as 
can be provided for the price they might be expected to pay, has given us 
an opportunity of controlling its developments such as rarely occurs. If all 
inventions were subject to similar delays, the control of the social effects of 
science would be much easier. As you know, a Committee is at work 
deciding how, if at all, television is to start. We hope they will not confine 
their attention to its start. When it starts, and if it succeeds at all, it is sure 
to develop in directions that we cannot at present foresee. The control of 
science, if it is to be effective, must be continuous and ever active. Let 
us hope that this exceptionally favourable chance will not be missed. 
The speed of photoelectric cells is also utilised in some other directions. 
Thus they are widely used in timing races, specially on greyhound tracks. 
Perhaps this is not an application that will appeal to you ; but after all, if 
greyhound races are to be timed at all, they may as well be timed rightly. 
Another possible application in the same direction may appeal to some of 
you still less. If a speed limit is imposed by the new Road Traffic Act, 
photoelectric cells might well be used to make it effective. The old system 
of timing over long distances by fallible constables armed with manual 
stop-watches is obsolete ; there would be no difficulty to-day in timing a car 
over a distance of 15 or 20 feet without possibility of human error and with 
all the accuracy required. 
The last class of applications, immensely varied, does not use either of 
the powers in which the cell surpasses the eye ; it uses merely the common 
power of distinguishing light from darkness. The only difference is then 
