ScpL 23, 1880] 



NATURE 



501 



colour, and having throughout a granulated or crystalline struc- 

 ture, and looking like a metal. In this form it is perfectly 

 opaque to light, even in very thin films. This variety of selenium 

 has long been known as " granular " or "crystalline " selenium, 

 or, as Regnault called it, "metallic " selenium. It was selenium 

 of this kind that Hittorff found to be a conductor of electricity 

 at ordinaiy temperatures. He alsa found that its resistance to 

 the passage of an electrical current diminished continuously by 

 heating up to the point of fusion, and that the resistance suddenly 

 increased in passing from the solid to the liquid condition. It 

 was early discovered that exposure to sunlight hastens the change 

 of selenium from one allotropic form to another ; and this 

 observation is significant in the light of recent discoveries. 



Although selenium has been kiiown for the last sixty years, it 

 has not yet been utilised to any extent in the arts, and it is still 

 considered simply as a chemical curiosity. It is usually supplied 

 in the form of cylindrical bars. These bars arc sometimes 

 found to be in the metallic condition ; Isut more usually they 

 are in the vitreous or non-conducting form. It occurred to 

 Willoughby Smith that, on account of tlie high resistance of 

 crystalline selenium, it might be usefully employed at the shore 

 end of a submarine cable, in his system of testing and signalling 

 during tlie process of submersion. Upon experiment, the 

 selenium was found to have all the resistance required — some of 

 the bars employed measuring as mucli as 1,400 megohms — a 

 resistance equivalent to that which would be offered Ijy a tele- 

 graph \vire long enough to reach from the earlli to the sun ! 

 But the resistance was found to be extremely variable. Experi- 

 ments were made to ascertain the cause of this variability. Mr. 

 May, Mr. Willoughby Smith's assistant, discovered that the 

 resistance was less when the selenium was exposed to liglit than 

 when it was in the dark. 



In order to be certain that temperature had nothing to do 

 with the effect, the selenium was placed in a vessel of water, so 

 that the light had to jsass through from 1 in. to 2 in. of water 

 in order to reach the selenium. The approach of a lighted 

 candle was found to be sufficient to cause a marked deflection of 

 the needle of the galvanometer connected with the selenium, 

 and the lighting of a piece of magnesium wire caused the 

 selenium to measure less than half the resistance it did the 

 moment before. 



These results l^■ere naturally at first received by scientific men 

 with some incredulity, but they w'ere verified by Sale, Draper, 

 Moss, and others. When selenium is exposed to the action of 

 the solar spectrum, the maximum effect is produced, according 

 to Sale, just outside the red end of the spectrum, in a point 

 nearly coincident with the maximum of the heat rays ; but, 

 according to Adams, the maximum eftect is produced in the 

 greenish-yellow or most luminous part of the specliiim. Lord 

 Rosse exposed selenium to the action of non-luminous radia- 

 tions from hot bodies, but could produce no effect ; whereas a 

 thermopile under similar circumstances gave abundant indications 

 of a current. He also cut off the heat rays from luminous 

 bodies by the interposition of liquid solutions, such as alum, 

 between the selenium and the source of light, A\ithout affecting 

 the power of the light to reduce the resistance of the selenium ; 

 whereas the interposition of these same substances almost com- 

 pletely neutralized the effect upon the thermopile. Adams 

 found that selenium was sensitive to the cold light of the moon, 

 and Werner Siemens discovered that, in certain extremely sensi- 

 tive varieties of selenium, heat and light produced opposite 

 effects. In Siemens' experiments, special arrangements were 

 made for the purpose of reducing the I'esistance of the selen- 

 ium employed. Two fine platinum wires were coiled together 

 in the shape of a double flat spiral in the zig-zag shape, 

 and were laid upon a plate of mica so that the discs did not 

 touch one another. A drop of melted selenium was then 

 placed upon the platinum wire arrangement, and a second 

 sheet of mica was pressed upon the selenium, so as to cause it to 

 spread out and fill the spaces between the wires. Each cell was 

 about the size of a silver dime. The selenium cells were then 

 placed in a paraffine bath, and exposed for some hours to a tem- 

 jjerature of 210 deg. C, after which they v.ere allowed to cool 

 \vith extreme slowness. The results obtained with these cells 

 were very extraordinary ; in some cases the resistance of the 

 cells, when exposed to light, was only one-fifteenth of their 

 resistance in the dark. 



W'ithout dwelling farther upon the researches of others, I may 

 say that the chief information concerning the effect of light 

 upon the conductivity of selenium will be found under the 



names of Willoughby Smith, Lieutenant Sale, Draper and 

 Moss, Professor W. G. Adams, Lord Rosse, Day, Sabini, 

 Dr. Werner Siemens, and Dr. C. W. Siemens. All observations 

 by these various authors had been made by means of galvano- 

 meters ; but it occurred to me that the telephone, from its ex- 

 treme sensitiveness to electrical influences, might be substituted 

 with advantage. Upon consideration of the subject, however, 

 I saw that the experiments could not be conducted in the 

 ordinary way for the following reason : — The law of audibility 

 of the teleplione is precisely analogous to the law of electric 

 induction. No eftect is produced during the passage of a con- 

 tinuous and steady current. It is only at the moment of change 

 from a stronger to a weaker state, or ■vice versa, that any audible 

 eftect is proposed, and tlie amount of effect is exactly proportional 

 to the amount of variation in the current. It was, therefore, 

 evident that the telephone could only respond to the effect pro- 

 duced in selenium at tlie moment of change from light to dark- 

 ness, or vice versa, and that it would be advisable to intermit the 

 light with great rapidity, so as to produce a succession of changes 

 in^the conductivity of the selenium, corresponding in frequency 

 to^music.il vibrations ^\'ithin the limits of the sense of hearing. 

 For I had often noticed that currents of electricity, so feeble as 

 to produce scarcely any audible effects from a telephone when 

 the circuit was simply opened or closed, caused very perceptible 

 musical sounds when the circuit was rapidly intenaipted, and 

 that the higher the pitch of sound the more audible was the 

 effect. I was much struck by the idea of producing sound by 

 the action of light in this way. Upon farther consideration it 

 appeared to me that all the audible effects obtained from varieties 

 of electricity could also be produced by variations of light acting 

 upon selenium. I saw that the effect could be produced at the 

 extreme distance at which selenium would respond to the action 

 of a luminous bod)', but that this distance could be indefinitely 

 increased by the use of a parallel beam of light, so that we 

 could telephone from one place to another without the necessity 

 of a conducting wire between the transmitter and receiver. 

 It was evidently necessary, in order to reduce this idea to 

 practice, to devise an apparatus to be operated by the voice of a 

 speaker, by which variations could be produced in a parallel 

 beam of light, corresponding to the variations in the air pro- 

 duced by the voice. 



I proposed to pass light through a large number of small 

 orifices, which might be of any convenient shape, but were 

 preferably in the form of slits. Two similarly perforated 

 plates were to be employed. One was to be fixed and the 

 other attached to the centre of a diaphragm actuated by the 

 voice, so that the vibration of the diaphragm would cause the 

 movable plate to slide to and fi'O over the surface of the fixed 

 plate, thus alternately enlarging and contracting the free orifices 

 for the passage of light. In this way the voice of a speaker 

 could control the amount of light passed through the perforated 

 plates without completely obstructing its passage. This appa- 

 ratus was to be placed in the path of a parallel beam of light, 

 and the undulatory beam emerging from the apparatus could be 

 received at some distant place upon a lens, or other apparatus, 

 by means of which it could be condensed upon a sensitive piece 

 of selenium placed in a local circuit with a telephone and 

 galvanic battery. The variations in the light produced by the 

 voice of the speaker should cause corresponding variations in 

 the electrical resistance of the selenium employed ; and the 

 telephone in circuit with it should reproduce audibly the tones 

 and articulations of the speaker's voice. I obtained some 

 selenium for the purpose of producing the apparatus shown, but 

 found that its resistance was almost infinitely greater than that 

 of any telephone that had been constructed, and I was unable to 

 obtain any audible eflects by the action of light. I believed, 

 however, that the obstacle could be overcome by devising mecha- 

 nical arrangements for reducing tke resistance of the selenium, 

 and by constructing special telephones for the purpose. I felt 

 so much confidence in this that, in a lecture delivered before the 

 Royal Institute of Great Britain, upon May 17, 1S7S, I announced 

 the possibility of hearing a shadow by interrupting the action of 

 light upon selenium. A few days afterwards my ideas upon this 

 subject received a fresh impetus by the announcement made by 

 Mr. Willoughby Smith before the Society of Telegraph En- 

 gineers that he had heard the action of a ray of light falling upon 

 a bar of crystalline selenium, by listening to a telephone in circuit 

 with it. 



It is not unlikely that the puMicity given to the speaking tele- 

 phone during the last few years may have suggested to many 



