Its evolution from the optical telegraph
It is common knowledge that the first railway semaphore was erected by Charles Hutton Gregory on the London and Croydon Railway (later the Brighton) at New Cross, southeast London, in the winter of 1842-1843 on the newly enlarged layout also accommodating the South Eastern Railway. These facilities had been inspected and approved by Maj-Gen C. W. Pasley, Inspecting Officer of the Board of Trade Railway Department. The semaphore was afterwards rapidly adopted as a fixed signal throughout Britain, superseding all others in most uses by 1870. The appearance of the first semaphore does not seem to have been recorded, but was surely similar to those frequently illustrated within the next decade, at London Bridge, Bricklayer's Arms Junction, and other neighbouring localities, and will be familiar to the reader. In this paper, I shall tell how the form of the semaphore was determined, and relate the central role played by General Pasley in this development.
The railway semaphore is an immediate descendant of the apparatus for optical telegraphs that were used from about 1793 until superseded by the electromagnetic telegraph, a process that was complete on land lines by 1850. Optical telegraphs survived at sea, and for ship-to-shore communication, until the appearance of radio. To appreciate the origin of the railway semaphore, it is necessary to say something about optical telegraphs. The invention of good telescopes about 1610 made optical telegraphs possible, reducing the number of observers required along the route. The challenge was to devise a method of showing signs that could be read at the greatest distances, and rapidly changed. Although there were suggestions, and mostly impractical ones, development was surprisingly slow. Robert Hooke's telegraph of 1684 displayed one of about 30 different distinctive shapes kept hidden behind a screen. The figure below shows a sketch of this apparatus.
The brothers Chappe in France, especially the elder, Claude, worked on a means of telecommunication for a number of years. They rejected the static electricity then known, and settled on a visual telegraph. By experiment, they determined the best form of display, which was the T-telegraph, consisting of a pivoted arm with two pivoted wings on its ends. This was painted black, and displayed with a sky background. It could be seen for over 10 miles, where other considerations allowed. In using the directional contrast between the parts of a line, the Chappes had hit upon a very sensitive property of vision. A line was built from Paris to the North in 1793 under the auspices of the National Assembly. When it brought news of a famous victory in a quarter of an hour, instead of the many hours required by a messenger on horseback, its future was assured. Later, a network of Chappe telegraphs covered France, and was a great favourite of Napoleon's. The name télégraphe, far-writer, was coined by Chappe. This work inspired a great number of inventors, not the least in Britain.
Richard Lovell Edgeworth (1744-1817) proposed a telegraph for Ireland when a French invasion was anticipated in 1794, and again in 1796, but it was not taken up. The rotating triangle of his tellograph is shown in the figure, as well as another suggestion for a two-armed telegraph. Gamble offered designs for a shutter telegraph and a 5-armed telegraph to the Admiralty in 1795, but the Admiralty finally adopted the Murray shutter telegraph for its lines to Portsmouth, Chatham and other coastal sites. None of these were nearly as visible as the Chappe T-telegraph.
How the Chappe telegraph was used is shown at the left, where two signs are shown passing from station to station. The sign changes when it is seen to be repeated at the next instrument. There was a special sign for repetition of the previous sign, since the sign had to change each time. Improvements on the Chappe telegraph were sought, chiefly along the line of sending more than one symbol at the same time. A complex 7-armed arrangement was erected on the roof of the Tuileries in 1796, but proved much too complicated. A modification, the three-armed French coast telegraph, appeared in 1803. This device was erected at a chain of observation posts around the coast to communicate with ships in the offing, and to report enemy activities. The chain of posts communicated with each other and with the Chappe telegraph by messengers, not by means of the signalling apparatus. This device was called the sémaphore, sign-carrier.
First Lt. Pasley, as he then was, began work on telegraphs while stationed in Malta. He planned to make communication more rapid by transmitting four symbols at a time, instead of only one. In the Chappe system, the more than 200 possible positions of the arms were individually interpreted by means of a code book at terminals. The telegraph operators had no idea of the information they were transmitting, which was very agreeable to the totalitarian French. In Britain, however, the semaphore was interpreted as giving numbers that referred to letters, numbers, entries in a dictionary, or some such. The modern practice of spelling out messages as used with the electromagnetic telegraph was considered too time-consuming. Col. Pasley's polygrammatic telegraph with four sets of two arms arranged side-by-side was devised in Malta in 1804. When he passed through England in 1807, he modified the telegraph to use a single vertical post on which the four pairs of arms were pivoted, one above the other. This invention was published in Tilloch's Philosophical Magazine, v. 29, p. 292 (1807). We should become better acquainted with General Pasley, since his work turned out to have much to do with railway signalling.
General Sir Charles William Pasley, KCB, FRS, DCL (Oxon.) (1780-1861) was a brilliant scholar, engineer and soldier, the founder of modern military engineering. He was born at Eskdalemoor, Dumfriesshire. At age 8 he read the bible in Greek, and at 12 composed a Latin history in the style of Livy dealing with schoolboy battles in his valley. He was educated at Selkirk, entering the Royal Military Academy, Woolwich in 1796, and commissioned 2nd Lt., Royal Artillery in 1797. Within a year, he transferred to the Royal Engineers, and served in the Mediterranean until 1807. He was at the sieges of Copenhagen (1807), Oviedo (1808, where his knowledge of Spanish helped Wellington), Walcheren and Flushing (1809). When wounded at the last engagement, he used his convalescence to learn German. He then commanded the engineers at Plymouth, in spite of his junior rank of Captain, for several years, and attracting favourable notice. He became the first director of field instruction at the new Royal Engineer Establishment at Chatham in 1812, promoted to brevet-Major, where he then remained for thirty years. The next year he became brevet-Lt Colonel, the next regimental Lt Colonel. He learnt Welsh and Irish from privates among the sappers and miners there. He was elected FRS in 1816, having already made a mark.
He became regimental Colonel in 1831. Underwater explosions were a speciality of his, as well as fortification, sieges, and optical telegraphy. He advocated uniform weights and measures and the decimal system, but opposed French (metric) units, which I find admirable and astute. He was promoted to Major General in 1841 when he became Inspector General of Railways, receiving his KCB when he stepped down from this position in 1846. He was, without a doubt, the most technically expert of any Railway Inspector, and set the Inspectorate off on the right foot. He became Lt General in 1851, and Colonel-Commandant of the Royal Engineers in 1853. In 1860, he was promoted to the highest rank, General. He lived at 12 Norfolk Crescent, Hyde Park, London.
Captain Pasley, as he then was, was not aware of the French coast telegraph until he left the Mediterranean, but it attracted his intense interest as soon as he became aware of it on the Walcheren expedition. He described it in detail in Tillotson's Philosophical Magazine, vol. 35, p. 339 (1810). Admiral Sir Home Popham devised land and ship semaphores that looked very much like Pasley's or the French coast semaphore, but quite simple and practical. One must appreciate that the signalling apparatus was not the only part of a telegraph; the inventors and promoters were at least as concerned with the way in which messages were transmitted, whether by code book, spelling, or other method. Nevertheless, there broke out some arguments over priority and so forth. Pasley was meticulous in giving credit to all contributors, and offended by unwarranted claims. Popham's semaphore was adopted by the Admiralty in 1816, replacing the cumbersome shutter telegraph on the line to Portsmouth. It does not seem to have been as rapidly adopted at sea, which is remarkable, but quite in line with the classic stupidity of the Admiralty at these times.
Pasley realised that the polygrammatic telegraph was too complicated, and required too many signalmen, and that simplicity was paramount. By 1822 he had designed his Universal Telegraph, which he intended for general use, not simply for military use in the field or aboard ship. An essential addition was the indicator, the short arm that projected from one side of the mast, giving a reference position. The various positions, differing by 45° , were numbered from 1 (on the side with the indicator) to 7. The telegraph could display positions 1 to 7, 12 to 17, 23 to 27, 34 to 37, 45 to 47, 56 to 57, and 67, plus Stop and Finished. Some positions were reserved to indicate alphabetic (17), numerical (35), or dictionary (26) interpretation, and repeat (4). This was a much simpler system than Home Popham's, less subject to misinterpretation, and more flexible. This telegraph was later much used at sea. The animation below shows the Universal Telegraph sending Pasley's name.
The most remarkable part of Pasley's proposal was for night signals. Four lamps were used, one at the pivot of the arms, one as an indicator at the same height but a distance away, and two at the ends of the arms. Pasley strongly recommended the use of oil lamps, not candles, for night signals. This is the first use of position-light aspects, and would have been quite practical. In a lithographed paper called Observations on Nocturnal Signals in General (Chatham: Royal Engineers, 1823), Pasley states that "Many attempts have been made from time to time to use coloured lights for night signals, all of which have failed, for the colour of a luminous point or line cannot be distinguished at any distance." This statement is of great importance in view of the later development of railway signals, in which colour became the principal means. When position-light railway signals were developed in 1915 on the Pennsylvania Railroad, five lamps in a row were used at first, later reduced to three, and the eye proved as sensitive to the orientation of the lights as it was to the directional contrast of the semaphore. The lights could be selected for maximum efficiency, so that the three lamps required less power than the single lamp of a colour-light signal. Astonishingly, these signals never were universally accepted, but have left a trace in Britain in the directional indicator of five white lights, used with colour-light signals. Pasley's original idea was, in fact, used in a few cases in the United States, and proved eminently practical.
The first models of the semaphore had rectangular arms, counterbalanced by a lead weight at the end of a latticed extension. They were to be one foot long for each mile of distance, and the width less than 2/13 but greater than 1/7 or 1/8 of the length. The arms, normally 5 to 6 feet long, were 'pannelled' with vertical slats, that could be made from light copper sheet. This style of arm, which persisted here and there until much later, is often said to have been made so to reduce wind resistance, but the true reason seems to be to make the arm as light as possible. The arms, placed outside the post front and back, were attached to sprocket wheels. Sprocket wheels were also attached to working levers at the base of the post, and the wheels were connected by an endless flat sprocket chain resembling that used on bicycles, giving positive motion. The arms were painted black and used with sky background, as was usual for semaphores. If a clear background was not available, the arms were to be painted white, or checquered, or in some contrasting fashion. The Atchison, Topeka and Santa Fe Railway, in the United States, much later, painted its semaphore arms in just this way, unlike all other railway companies in the United States or Britain.
Very soon, the construction of the semaphore was altered by Mr Robert Howe, Clerk of Works at the Royal Engineer Establishment. As appearing in the Field Instructions for the Royal Engineers, p. 36, the arms were now placed within the signal post, a simpler construction that made the bearings much easier to fabricate. The arms shown were tapered, oddly, to be narrower at the ends than at the pivot, but it was stated that Admiralty semaphores had parallel blades. It takes very little imagination to notice that this is precisely the construction of the early railway semaphores. General Pasley suggested the semaphore to Mr Gregory as a reliable means to communicate with the driver, more certain than the flag signals previously used, and able to be seen at a greater distance. All Gregory had to do was set one up and decide how it should be used. Night signals were not shown as suggested by the General, but by hanging coloured lamps on the semaphore post. The further development of the semaphore is part of railway signal engineering history, but this is how it came about.
There is a very good book about optical telegraphs by Geoffrey Wilson, The Old Telegraphs (London: Phillimore, 1976). The suggestion there that the railway semaphore was suggested by a telegraph entrepreneur with a station in the vicinity is extremely improbable. I lack biographical information on Sir Charles Hutton Gregory, who for some inexplicable reason does not even appear among the 'missing persons' of the National Dictionary of Biography, even though he was President of the Institution of Civil Engineers. From his name, I would assume he is a Scot, like Pasley. Some work could be done to find out if there is any record of the interaction of Gregory and Pasley in the 1840's, perhaps in parliamentary papers. General Pasley's own account of optical telegraphs appears in the References.
Pasley's proposal to use optical telegraphs observed by locomotive drivers for railway signals was quite original. In Germany, two-arm optical telegraphs were used, and were the ancestors of semaphores there, but for many years were only for communication between employees on the ground. In Britain (and the U.S.) only rotating targets, suspended balls, and devices suggesting flags were used. A good example were Hall's "panel" signals on the Eastern Counties Railway, also introduced in 1842 (See the References). Most of the early fixed signals were either station signals to protect trains in a station or at junctions, or time-interval signals. The purpose of time-interval signals was to ensure that there was adequate time for a flagman to protect a train stopped out of course. Their indications were to stop, or to proceed slowly, if a train had just passed. There were many patented railway signals, most of which were impractical and never appeared in practice. Many were operated by the train itself, and had to be replaced manually. Such signals were used occasionally in early days in Britain, but were completely superseded by block working. Block working, depending on the electric telegraph, was introduced slowly after 1851 in Britain, but did not arrive in America until 1876.
C. W. Pasley, LtCol RE, FRS, Description of the Universal Telegraph for Day and Night Signals (London: 1823; BL Shelfmark T.1105.).
C. Cock, Hall's Panel Signals on the Eastern Counties Railway, The Signalling Record, No. 122, Mar/Apr 2007, pp. 39-45.
Composed by J. B. Calvert
Created 15 April 2000
Last Revised 4 May 2007