122 



NA TURE 



[May 12, 1910 



respectively. Dr. Smart, using Pontecoulant's elements, 

 obtained 3i'3 and 062 miles per second. 



Messrs. Cowell and Crommeiin have been awarded, 

 jointly, the Janssen medal of the Soci6t^ astronomique de 

 France, for their precise determination of the orbit of the 

 comet for this present apparition. 



.\ number of interesting representations of comets, some 

 certainly of Halley's, are reproduced in the May number 

 of the Bulletin de la Society astronomique de France from 

 the " Theatrum Cometicum " of Lubienietz. Each draw- 

 ing is accompanied by a note explaining it, and directing 

 attention to contemporary occurrences ; in concluding the 

 article, M. Flammarion suggests that great comets were 

 of more frequent occurrence in early times than they are 

 now. 



.4 NEW TELEPHONE RELAY AND ITS 

 APPLICATIONS.^ 



"pVER since the introduction of the telephone a real 

 •*-' need was felt for a telephone relay, for the distance 

 over which telephones could be used was found to be com- 

 paratively limited. Edison, soon after his invention of the 

 carbon button transmitter, caused an electromagnet to act 

 upon the iron diaphragm, and thus turned it into a relay, 

 but it was not a success. Hughes (Proceedings of the 

 Royal Society, vol. xxvii., p. 362, 1878), in his paper before 

 the Royal Society in 1878, describing his extremely delicate 

 microphones, stated that a telephone receiver, if included 

 in the microphone circuit and placed upon the resonant 

 board, caused a continuous sound to be produced. It 

 follows, he said, that the question of providing a relay for 

 the human voice in telephony is thus solved. Unfortu- 

 nately, it was not solved; he had shown how to make a 

 relay that would magnify a noise or musical note, but not 

 one that would intensify articulate speech. 



Sir Oliver Lodge (Journal of the Institution of Electrical 

 Engineers, vol. xxvii., p. 799, 1898), in a paper read in 

 December, 1898, before this society, described a relay con- 

 sisting of three or four reeds or tuning-forks, each carry- 

 ing carbon contacts and working in series with one another. 

 Each reed was arranged to resonate to one particular 

 musical note, and when this note was passed through the 

 string of relays it was multiplied in power to a considerable 

 extent. An instrument of this character, however, is not 

 effective in intensifying speech. An articulate relay must 

 hav^e its vibrating parts damped, or, in other words, possess 

 no resonating properties ; it is therefore far more insensitive 

 to sound than one that is arranged to resonate to one 

 particular note. 



The invention of the powerful granular transmitters of 

 the Hunning type stimulated further efforts to obtain the 

 speaking relay, and some progress was made with this 

 type of microphone, particularly in America. I will not 

 describe these relays further than to say that they consist 

 in combining the telephone receiver and the granular 

 carbon transmitter ; both of these are designed as efficiently 

 as possible, and in some cases automatic means are pro- 

 vided to shake up the granules should they become packed. 

 These relays are onh* partially successful. Their 

 advantages are not decisive. They require relatively 

 powerful currents to work them ; that is to say, when the 

 telephone currents become sufficienth' feeble to require 

 their services, it is at this point that the carbon instru- 

 ment fails to work. The telephone relay to be successful 

 has to magnify in a continuous manner varying currents 

 that are .too feeble to affect properly a Bell teleohone 

 receiver. Such currents would be of excessive weakness, 

 say of the order of the one one-hundred millionth of an 

 ampere (10-' ampere), and the mechanical movements pro- 

 duced by such currents, which have in their turn to bring 

 about the increased electrical changes, are therefore micro- 

 scopic in dimensions. 



The author's telephone relay has had to be developed 

 along quite new lines. It takes as its basis the researches 

 of J- J- Thomson, Earhart, Kinsley, and others, with 

 regard to the flow of electrons across a microscopic air- 

 gap between two conducting surfaces at different potentials 

 (see " Conduction of Electricity through Gases," J. J. 



^ From a paper read Vefore the Instituiion of Electrical Engireers on 

 May 5 by Mr. S. G. Brown. 



Thomson, chap. xv.). Earhart made a series of experi- 

 ments on the difference of potential required to produce 

 sparks the length of which is comparable with the wave- 

 length of sodium light, and he found that when the distance 

 between the metal electrodes falls to less than about 

 3x10-* cm., the spark potential falls off rapidly with 

 the distance, and seems to become proportional to the 

 distance ; that is to say, when the electrodes are placed 

 ver\- close together, within a distance such that the average 

 intensity of force F between the electrodes reaches a value 

 of about a million volts per centimetre, the discharge or 



NO. 21 15, VOL. 83] 



Fig. I. 



current passing is determined by the condition that F, 

 which is \'ld, reaches this value (where V is the potential • 

 difference and d the distance between the electrodes). If : 

 the metallic circuit of a dry cell be interrupted by a minute 

 opening or space of the order of 5x10-' cm., the metal 

 at the point of interruption being platinum, the current will 

 continue to flow round the circuit and across the open- 

 ing, and any slight alteration in the length of the space, ; 

 which I shall call the conduction space, will vary its, 

 resistance and greatly affect the value of the current that ; 

 flows round the circuit. This conduction space is there- 

 fore exactly what is wanted for the current-varying device 

 of a telephone relay, 

 where microscopic 



mechanical m o v e- 

 ments are to be 

 transformed into 

 large current changes. 



The dimensions of the _ v^l-^ 



conduction space are Oil — rrV^^l-^0 



so small that it is K\\\\\\\\\\\\\\\\X\\ V\\\\^\'^^ 

 difficult to ensure and 

 maintain it by direct . 

 mechanical Vi e a n s. 

 The current- that 

 flows across the space 

 was therefore made to 

 do its own adjust- 

 ment, very much in 

 the same way as the 

 current that passes through the arc of an arc lamp is made 

 to strike and maintain the length of the arc. 



Fig. I is a side view of the instrument with the brass 

 cover removed. N is a permanent magnet, continued by 

 soft iron poles right up to, but not touching, the " invar " 

 steel reed P. Round the soft iron pole extensions are 

 wound the two sets of coil windings H and K. The tele- 

 phone currents to be magnified circulate round the wind- 

 ing H, and thus, by varying the magnetism, set the reed 

 P in vibration. M, O are the top and bottom metal 

 contact-pieces, which are opened to an infinitesimal degree 

 to form a microphone by the fine adjusting screw W and 



