744 
NATURE 
= 
[JuNE 2, 1923 
The Transmission of Speech by Light. 
By Prof. A. O. RANKINE. 
‘a 1880 Graham Bell devised a system of using light 
for transmitting sounds, including speech, and 
called his instrument the “ photophone.”’ This system 
afterwards experienced a chequered career, having 
attracted only occasionally the attention of other 
investigators, with the result that, although consider- 
able improvements have been made, it has until quite 
recently remained a novelty. It is beginning, however, 
to enter upon the phase of practical use, more par- 
ticularly in connexion with some of its applications in 
which the distance over which the light acts as the 
vehicle plays no essential part. In these circumstances 
it is, perhaps, desirable to introduce some modification 
of nomenclature. There is little doubt that Graham 
Bell’s original idea was to transmit speech by means of 
a beam of light which travelled over as great a distance 
as possible. It is true that the maximum range he 
records having attained is 7oo feet. The modesty of 
this achievement perhaps prevented him from. in- 
troducing the notion of distance into the name he gave 
to the apparatus. Now, however, the name “ photo- 
telephony ” would appear to be appropriate for the 
improved system which, with its increased efficiency, 
and with the aid of modern amplifying devices, has a 
range of transmission of several miles and the im- 
mediate prospect of such extension that the earth’s 
curvature will prove in practice to be the limiting 
factor. The adoption of this name for telephony by 
light would have the additional advantage that it 
would leave us free to retain the equally appropriate 
term “ photophone” for those special modifications, 
already mentioned, in which distance is an unimportant 
consideration. 
If we wish to use light for the transmission of sounds 
it is clear that we must impose on the light features 
which are characteristic of the sounds in question. 
The plan generally adopted is to modulate the intensity 
of the light in accordance with the vibrations con- 
stituting the sounds. How this is done will be con- 
sidered later. For the moment the question is by 
what means these fluctuations of intensity can be made 
to reproduce audibly the original vibrations. This 
reproduction is possible because we have at our disposal 
certain substances, of which suitably prepared selenium 
is the best-known example, capable of acting as electric 
valves operated by variations of illumination. Selenium 
is not ideal for the purpose. Indeed, having regard to 
its many defects, it is surprising that it functions so 
well as it does. Shelford Bidwell, in a Friday evening 
discourse at the Royal Institution in 1881, spoke of 
the ‘“‘capricious behaviour” of selenium, and it has 
to be admitted that this is still a fair description, even 
though many improvements of design and efficiency 
have been introduced by various makers of so-called 
selenium cells. There are, for example, obscure changes 
of conductivity, occurring slowly, which have defeated 
all attempts to use the electrical conduction of selentum 
as a basis in photometry. 
Fortunately, these relatively slow changes are not 
of appreciable importance in connexion with the 
aaa two lectures delivered at the |Royal Institution on April 12 
and 19. 
NO. 2796, VOL. 111 | 


photophone, for in that case the fluctuations of light 
intensity are very rapid, corresponding, as they do, 
to audible frequencies. A more objectionable feature 
is that there is displayed a considerable lag in the 
electrical valve action, which prevents its operation — 
being at all efficient, especially when the selenium is 
called upon to respond, as in the photophone, many 
times a second. If a suitable substitute free from this 
inertia-like effect could be found it would very soon 
displace selenium cells and the similar devices at 
present in use. Of the latter a notable example is the 
“ thalofide ”’ cell of T. W. Case, which is quicker than 
selenium in its response, but is sensitive to infra-red 
radiation rather than to visible light, and cannot be 
exposed to bright light without suffering deterioration 
in its photo-electric properties. : 
It is not proposed here to describe in detail the 
transmitting and receiving devices employed in speech 
transmission by light. The writer has, in a previous 
issue of NATURE,” already given some indication of the 
lines of development. With regard to the modulation 
of the light by speech two general plans have been 
adopted, namely (1) to cause the speech vibrations 
to control the actual candle-power of an artificial 
source of light, and (2) to use the voice to actuate a 
mechanism which interrupts in the appropriate manner 
a beam of light after it has left a constant source. 
Graham Bell’s transmitter was of the latter type, and, 
although until recently the tendency, particularly on 
the continent, has been to employ the former plan by 
superimposing on the current in an electric arc or 
suitable filament lamp the microphonic currents arising 
from speech sounds, it is now fairly generally recognised 
that greater efficiency can be attained by improved 
forms of the interruption method. This method has 
also the advantage that it can be applied to any source 
of light, and thus brings into our service sunlight, which 
is the brightest of all. 
The simplest form of the receiving device is a circuit 
consisting of a selenium cell, an electric battery, and a 
telephone receiver. On exposure of the selenium to 
constant illumination, a constant, or, at any rate, a 
very slowly varying current passes. If, however, the 
illumination is of a fluctuating character—if, in par- 
ticular, the variations are those due to modulation by 
speech vibrations—the selenium is able, in spite of its 
lag, to control the current in a closely corresponding 
manner, so that the diaphragm in the telephone 
receiver, through which this current flows, is set in 
vibration, and emits sounds resembling with a re- 
markable degree of accuracy the original sounds used 
in modulating the light. The speech currents in this 
simple circuit can be transformed into other circuits: 
if desired, and they can be amplified in the usual way 
by means of thermionic valves. It only has to be 
borne in mind that selenium cells are usually of very 
high resistance, and that, therefore, methods of trans- 
formation appropriate to such cases should be employed. 
It is easy to arrange the optical system of the 
transmitter so that the light projected is confined to a 
very narrow angle and directed upon any small chosen 
* Nature, vol, 104, p.{604 (1920). 

