T. A. EDISON-1885.

Electric communication with trains in motion, like communication with ships at sea and with lighthouses, has long been a favourite problem with electrical engineers: indeed it is much the older of the two, and dates back to the first days of electric telegraphy.

In 1838 Edward Davy, the rival of Cooke and Wheatstone, proposed such a system. In a lecture on "Electric Telegraphy," delivered in London during the summer of 1838, he says:

"I have a few words to say upon another application of electricity—namely, the purposes it will answer upon a railway, for giving notices of trains, of accidents, and stoppages. The numerous accidents which have occurred on railways seem to call for some remedy of the kind; and when future improvements shall have augmented the speed of travelling to a velocity which cannot at present be deemed safe, then every aid which science can afford must be called in to promote this object. Now, there is a contrivance, secured by patent, by which, at every station along the railway line, it may be seen by mere inspection of a dial what is the exact situation of the engines running either towards or from the station, and at what speed they are travelling. Every time the engine passes a milestone, the pointer on the dial moves forward to the next figure, a sound or alarm accompanying each movement.

"Not only this, but if two engines are approaching each other, by any casualty, on the same rails, then, at a distance of a mile or two, a timely notice can be given in each engine by a sound or alarm, from which the engineer would be apprised to slacken the speed; or, if the engineer be asleep or intoxicated, the same action might turn off the

steam, independently of his attention, and thus prevent an accident." >> 1

In 1842 William Fothergill Cooke published his ‘Telegraphic Railways,' descriptive of a crude system of train signals, which was tried, in 1843, in the Queen Street tunnel, Glasgow, and in the Clay Cross tunnel, Derby; and, on a more extensive scale, in 1844, on the Great Eastern Railway, between Norwich and Yarmouth.

Dujardin in 1845, Brett and Little in 1847, Edwin Clark in 1854, Bonelli in 1855, and many others, proposed various systems of train signalling; but as they are all based on ordinary telegraphic principles and require connecting wires, they do not specially concern us in this history.

Mr A. C. Brown, an officer of the Eastern Telegraph Company, claims to have been the first to suggest, in 1881, the method of induction for communicating with moving trains. In a letter published in the 'Electrician,' March 21, 1885, he says:

"My object was chiefly to provide an efficient means of fog-signalling, by enabling the signalman to communicate directly with the drivers or guards. I proposed to run a wire along the permanent way, parallel with the rails, and to wind a coil of wire round the engine, or carriage to be communicated with, in such a way as to get as long a length of wire parallel to, and as near to, the line-wire as possible, so as to be well exposed to the inductive action thereof. then proposed to place in the signal-boxes a battery, signalling key, and rapid make-and-break instrument, or buzzer, and to thereby signal to the train, using a telephone in circuit with the train-coil as a receiver. By using an ordin

I

1 See the writer's 'History of Electric Telegraphy,' p. 407. The most perfect block system of the present day does not do anything like this. Davy's plan was actually patented by Henry Pinkus! See his patent specification, No. 8644, of September 24, 1840.

ary carbon transmitter in the line-wire, I also found it quite practicable to speak verbally to the train, so as to be distinctly heard in the telephone.

"This design was embodied in a paper which, in the year 1881, I laid before the managing director of the United Telephone Company, but want of time and opportunity prevented its being put into practice. It was experimentally tried at that time, using wire coils, properly proportioned in length, resistance, and distance apart to the conditions that would be obtained in practice. It has since been simplified and arranged to produce both visible and audible signals on the engine or car by induction from a No. 8 iron line-wire across a space of 6 inches, with a current of only one quarter ampère, or such as can easily be produced by the ordinary Daniell batteries used in railway work." 1

In 1883 Mr Willoughby Smith threw out a similar suggestion towards the end of his paper on "Voltaic-Electric Induction," read before the Institution of Electrical Engineers, November 8 of that year: 2

"Telegraph engineers," he says, "have done much towards accomplishing the successful working of our present railway system, but still there is much scope for improvements in the signalling arrangements. In foggy weather the system now adopted is comparatively useless, and recourse has to be had at such times to the dangerous and somewhat clumsy method of signalling by means of detonating charges placed upon the rails.

"Now, it has occurred to me that Volta-Electric induction might be employed with advantage in various ways for signalling purposes. For example, one or more spirals could be fixed between the rails at any convenient distance

1 For another proposal of Mr Brown, see p. 175, infra.

2 Compare also his remarks, 'Jour. Inst. Elec. Engs.,' March 23, 1882, p. 144.

from the signalling station, so that, when necessary, intermittent currents could be sent through the spirals; and another spiral could be fixed beneath the engine, or guard's van, and connected to one or more telephones placed near those in charge of the train. Then, as the train passed over the fixed spiral, the sound given out by the transmitter would be loudly reproduced by the telephone, and indicate by its character the signal intended.

"One of my experiments in this direction will perhaps better illustrate my meaning. The large spiral was connected in circuit with twelve Leclanché cells and the two make-and-break transmitters before described. They were so connected that either transmitter could be switched into circuit when required, and this I considered the signalling station. The small spiral was so arranged that it passed in front of the large one at the distance of 8 inches, and at a speed of twenty-eight miles per hour. The terminals of the small spiral were connected to a telephone fixed in a distant room, the result being that the sound produced from either transmitter could be clearly heard and recognised every time. the spirals passed each other. With a knowledge of this fact I think it will be readily understood how a cheap and efficient adjunct to the present system of railway signalling could be obtained by such means as I have ventured to bring to your notice this evening."

In 1885 Mr T. A. Edison had his attention directed to the subject, and with his usual thoroughness he soon produced a very complete system, with the assistance of Messrs Gilliland, Phelps, and W. Smith to the lastnamed of whom the original idea is said to be due.1

1 Although I have not seen any acknowledgment of their indebtedness, Mr Edison and his coadjutors can hardly have been ignorant of Mr Willoughby Smith's very clear proposal, of which their contrivance is but the practical realisation. Given the idea, the rest was easy enough.

The inevitable avant-coureur appeared in the technical journals of the period, and as it is delightfully characteristic of the great magician of Menlo Park, we venture to reproduce it here: "Mr Edison's latest invention, an arrangement to telegraph from moving trains, is thus described by a recent visitor to his laboratory: Overhead was a board eight inches wide, suspended from the ceiling by ropes fastened to one of its edges. One side of it was covered with tinfoil, and was facing toward a wall 20 feet distant. That,' said Mr Edison, is my railroad signal; I make electricity jump 35 feet, and carry a message. This is something quite new; no induction has ever been known that extended over 3 or 4 or 5 feet. This invention uses what is called static electricity, and it makes every running train of cars a telegraph station, accessible to every other telegraph station on the road. Messages may be sent to and from conductors, and to and from passengers. It requires no extra wire, either under the cars or at the side of the cars, but uses the ordinary telegraph just as it is put up at the side of the track. This white board is a receiver and transmitter. A board like it is to be fastened lengthwise along the peak of each car, where it will be out of the way and will not be a blemish. When the train is telegraphed to, the message jumps from the wire on the side of the track and alights on this board, and is conveyed to the apparatus in the train below. It works beautifully from those wires strung yonder. I was as much astonished as anybody at finding out what could be done. It costs very little, moreover, as 300 miles of road can be equipped for 1000 dols.'

This contrivance was patented in England on June 22, 1885, in the joint names of T. A. Edison and E. T. Gilliland, and is fully described in their specification, No. 7583, of which the following is an abstract :

--