View source for Engineering:Optical telegraph

{{Short description|Communication along a chain of towers using mechanically operated paddles or shutters}}
[[File:OptischerTelegraf.jpg|thumb|upright|A replica of one of [[Biography:Claude Chappe|Chappe's]] semaphore towers in Nalbach, Germany]]
[[File:Signaling by Napoleonic semaphore line.jpg|thumb|Illustration of signalling by semaphore in 18th-century France. The operators would move the semaphore arms to successive positions to spell out text messages in semaphore code, and the people in the next tower would read them.]]

An '''optical telegraph''' is a line of stations, typically towers, for the purpose of conveying textual information by means of visual signals (a form of [[Engineering:Optical communication|optical communication]]).  There are two main types of such systems; the '''semaphore telegraph''' which uses pivoted indicator arms and conveys information according to the direction the indicators point, and the '''shutter telegraph''' which uses panels that can be rotated to block or pass the light from the sky behind to convey information.

The most widely used system was the [[Engineering:Chappe telegraph|Chappe telegraph]], which was invented in France in 1792 by [[Biography:Claude Chappe|Claude Chappe]]. It was popular in the late eighteenth to early nineteenth centuries.<ref name="Burns2004">{{harvnb|Burns|2004|loc=[https://books.google.com/books?id=7eUUy8-VvwoC&pg=PA29 Chapter 2: Semaphore Signalling]}}</ref><ref name="EncyBrit1824">{{cite book |chapter-url=https://books.google.com/books?id=MsYnAAAAMAAJ&pg=PA645 |chapter=Telegraph |volume=10 |title=Encyclopædia Britannica |edition=6th |year=1824 |pages=645–651 }}</ref><ref name="EdEnc1832">{{cite book |chapter-url=https://books.google.com/books?id=VhEbAQAAMAAJ&pg=PA657 |chapter=Telegraph |volume=17 |title=The Edinburgh Encyclopaedia |pages=664–667 |year=1832 |editor=David Brewster }}</ref> Chappe used the term "télégraphe" to describe the mechanism he had invented – that is the origin of the English word "telegraph".<ref name="Axon1880">{{cite journal |last1=Axon |first1=William |title=On the History of the word Telegraph |journal=Proceedings of the Literary and Philosophical Society of Manchester |date=1880 |volume=19 |page=183 |url=https://books.google.com/books?id=h_0NAAAAYAAJ&pg=RA3-PA183 |access-date=10 June 2023}}</ref> Lines of relay towers with a semaphore rig at the top were built within line of sight of each other, at separations of {{convert|5-20|mi|km|0}}. Operators at each tower would watch the neighboring tower through a [[Astronomy:Telescope|telescope]], and when the semaphore arms began to move spelling out a message, they would pass the message on to the next tower.  This system was much faster than post riders for conveying a message over long distances, and also had cheaper long-term operating costs, once constructed. Half a century later, semaphore lines were replaced by the electrical telegraph, which was cheaper, faster, and more private. The line-of-sight distance between relay stations was limited by [[Earth:Geography|geography]] and weather, and prevented the optical telegraph from crossing wide expanses of water, unless a convenient island could be used for a relay station. A modern derivative of the semaphore system is [[Engineering:Flag semaphore|flag semaphore]], signalling with hand-held flags.

==Etymology and terminology==
The word ''semaphore'' was coined in 1801 by the French inventor of the semaphore line itself, [[Biography:Claude Chappe|Claude Chappe]].<ref>''Groundbreaking Scientific Experiments, Inventions & Discoveries of the 18th Century'', Jonathan Shectman, p. 172</ref> He composed it from the [[Social:Greek language|Greek]] elements σῆμα (sêma, "sign"); and from φορός (phorós, "carrying"),<ref>''Oxford English Dictionary''.</ref> or φορά (phorá, "a carrying") from φέρειν  (phérein, "to bear").<ref>''Webster's Unabridged Dictionary''.</ref> Chappe also coined the word ''tachygraph'', meaning "fast writer".<ref>Beyer, Rick, ''The Greatest Stories Never Told'', A&E Television Networks / The History Channel, {{ISBN|0-06-001401-6}}, p. 60</ref> However, the French Army preferred to call Chappe's semaphore system the ''telegraph'', meaning "far writer", which was coined by French statesman André François Miot de Mélito.<ref>Le Robert historique de la langue française, 1992, 1998</ref> 

The word ''semaphoric'' was first printed in English in 1808: "The newly constructed Semaphoric telegraphs (...) have been blown up", in a news report in The Naval Chronicle.<ref>{{Cite book |last=Sherk |first=Bill |url=https://www.academia.edu/28929229 |title=500 Years of New Words |publisher=The Dundurn Group |year=2004 |isbn=1-55002-525-2 |location=Toronto |pages=210}}</ref> The first use of the word ''semaphore'' in reference to English use was in 1816: "The improved Semaphore has been erected on the top of the Admiralty",<ref>{{Cite book |last=Sherk |first=Bill |url=https://books.google.com/books?id=E8t3hZgrVxkC&dq=%2522The+improved+Semaphore+has+been+erected+on+the+top+of+the+Admiralty%2522&pg=PA210 |title=500 Years of New Words: the fascinating story of how, when, and why these words first entered the English language |date=2004-09-01 |publisher=Dundurn |isbn=978-1-4597-1822-7 |pages=320 |language=en}}</ref><ref>{{cite book |chapter-url=https://www.british-history.ac.uk/survey-london/vol16/pt1/pp45-70 |title=Survey of London |chapter=The Admiralty |editor1-first= G. H. |editor1-last=Gater |editor2-first=E. P. |editor2-last=Wheeler |year=1935 |volume=16 |pages=45–70 |publisher=London County Council |via=British History Online}}</ref> referring to the installation of a simpler telegraph invented by Sir Home Popham.{{citation needed|date=August 2016}} Semaphore telegraphs are also called, "Chappe telegraphs" or "Napoleonic semaphore".<ref>[http://www.saintefoyleslyon.fr/index.php?rubrique=142 Tour du télégraphe Chappe] {{in lang|fr}} {{webarchive|url=https://web.archive.org/web/20110928210813/http://www.saintefoyleslyon.fr/index.php?rubrique=142 |date=28 September 2011 }}</ref><ref>{{Cite news |last=Schofield |first=Hugh |date=2013-06-16 |title=How Napoleon's semaphore telegraph changed the world |language=en-GB |work=BBC News |url=https://www.bbc.com/news/magazine-22909590 |access-date=2022-03-03}}</ref>

==Early designs==
[[File:Rees's Cyclopaedia - Hooke's optical telegraph.png|thumb|Illustration showing Robert Hooke's proposed system. At top are various symbols that might be used; ABCE indicates the frame, and D the screen behind which each of the symbols are hidden when not in use.]]
Optical telegraphy dates from ancient times, in the form of [[Engineering:Hydraulic telegraph|hydraulic telegraph]]s, torches (as used by ancient cultures since the discovery of fire) and smoke signals. Modern design of semaphores was first foreseen by the British [[History:Polymath|polymath]] Robert Hooke, who gave a vivid and comprehensive outline of visual telegraphy to the Royal Society in a 1684 submission in which he outlined many practical details. The system (which was motivated by military concerns, following the Battle of Vienna in 1683) was never put into practice.<ref>{{cite web|url=http://mysite.du.edu/~jcalvert/railway/semaphor/semhist.htm |title=The Origin of the Railway Semaphore |publisher=Mysite.du.edu |access-date=2013-06-17}}</ref><ref>{{cite web |url=http://www.ilt.columbia.edu/projects/bluetelephone/html/part2.html |title=History of the Telephone part2 |publisher=Ilt.columbia.edu |access-date=2013-06-17 |archive-url=https://web.archive.org/web/20121128120005/http://www.ilt.columbia.edu/projects/bluetelephone/html/part2.html |archive-date=2012-11-28 |url-status=dead }}</ref>

[[File:Rees's Cyclopaedia - Richard Lovell Edgeworth's optical telegraph.png|thumb|Sir Richard Lovell Edgeworth's proposed optical telegraph for use in Ireland. The rotational position of each one of the four indicators represented a number 1-7 (0 being "rest"), forming a four-digit number. The number stood for a particular word in a codebook.]]
One of the first experiments of optical signalling was carried out by the Anglo-Irish landowner and inventor, Sir Richard Lovell Edgeworth in 1767.<ref>{{cite encyclopedia|editor-first=Abraham|editor-last=Rees|editor-link=Abraham Rees|encyclopedia=Cyclopædia|title=Telegraph|url=https://archive.org/details/cyclopaediaorun35rees|year=1802–1820|publisher=Longman, Hurst, Rees, Orme & Brown
|volume=35|location=London|at=Unpaginated work: pages 9-11 of the article entry}}</ref> He placed a bet with his friend, the horse racing gambler Lord March, that he could transmit knowledge of the outcome of the race in just one hour. Using a network of signalling sections erected on high ground, the signal would be observed from one station to the next by means of a [[Astronomy:Telescope|telescope]].{{sfn|Burns|2004}} The signal itself consisted of a large pointer that could be placed into eight possible positions in 45 degree increments. A series of two such signals gave a total 64 code elements and a third signal took it up to 512. He returned to his idea in 1795, after hearing of Chappe's system.

==Prevalence==

===France===
[[File:Chappe semaphore.jpg|thumb|upright|19th-century demonstration of the semaphore]]
Credit for the first successful optical telegraph goes to the French engineer [[Biography:Claude Chappe|Claude Chappe]] and his brothers in 1792, who succeeded in covering France with a network of 556 stations stretching a total distance of {{convert|4,800|km}}. ''Le système Chappe'' was used for military and national communications until the 1850s.<!-- Another source says 534 stations and more than 5,000km. https://www.bbc.co.uk/news/magazine-22909590 -->

====Development in France<!-- "France" required to avoid repeated heandin name per MOS:HEAD -->====
During 1790–1795, at the height of the {{wipe|French Revolution}}, France needed a swift and reliable [[Social:Military communications|military communications]] system to thwart the war efforts of its enemies. France was surrounded by the forces of Britain, the Netherlands, [[Place:Prussia|Prussia]], Austria, and Spain, the cities of Marseille and Lyon were in revolt, and the British Fleet held Toulon. The only advantage France held was the lack of cooperation between the allied forces due to their inadequate lines of communication. In mid-1790, the Chappe brothers set about devising a system of communication that would allow the central government to receive intelligence and to transmit orders in the shortest possible time. Chappe considered many possible methods including audio and smoke.  He even considered using electricity, but could not find insulation for the conductors that would withstand the high-voltage [[Physics:Electrostatic generator|electrostatic sources]] available at the time.<ref>Patrice Flichy, ''Dynamics of Modern Communication'', p. 33, SAGE, 1995 {{ISBN|144622712X}}</ref><ref>Holzmann & Pehrson, p. 53</ref>

Chappe settled on an optical system and the first public demonstration occurred on 2 March 1791 between Brûlon and Parcé, a distance of {{convert|16|km}}. The system consisted of a modified pendulum clock at each end with dials marked with ten numerals.  The hands of the clocks almost certainly moved much faster than a normal clock.  The hands of both clocks were set in motion at the same time with a synchronisation signal.  Further signals indicated the time at which the dial should be read.  The numbers sent were then looked up in a [[Social:Codebook|codebook]].  In their preliminary experiments over a shorter distance, the Chappes had banged a pan for synchronisation.  In the demonstration, they used black and white panels observed with a telescope.  The message to be sent was chosen by town officials at Brûlon and sent by René Chappe to Claude Chappe at Parcé who had no pre-knowledge of the message.  The message read "si vous réussissez, vous serez bientôt couverts de gloire" (If you succeed, you will soon bask in glory). It was only later that Chappe realised that he could dispense with the clocks and the synchronisation system itself could be used to pass messages.<ref>Holzmann & Pehrson, pp. 53–55</ref>

The Chappes carried out experiments during the next two years, and on two occasions their apparatus at Place de l'Étoile, [[Place:Paris|Paris]] was destroyed by mobs who thought they were communicating with royalist forces. Their cause was assisted by Ignace Chappe being elected to the Legislative Assembly.  In the summer of 1792 Claude was appointed ''Ingénieur-Télégraphiste'' and charged with establishing a line of stations between Paris and Lille, a distance of 230 kilometres (about 143 miles). It was used to carry dispatches for the war between France and Austria. In 1794, it brought news of a French capture of Condé-sur-l'Escaut from the Austrians less than an hour after it occurred.<ref>[https://www.bbc.co.uk/news/magazine-22909590 How Napoleon's semaphore telegraph changed the world], BBC News, Hugh Schofield, 16 June 2013</ref><!-- Note: this BBC source tells this story well, and supports the Lille and Conde statements, but was not used for the whole section, which remains 'unreferenced' --> The first symbol of a message to Lille would pass through 15 stations in only nine minutes. The speed of the line varied with the weather, but the line to Lille typically transferred 36 symbols, a complete message, in about 32 minutes. Another line of 50 stations was completed in 1798, covering 488&nbsp;km between Paris and Strasbourg.<ref>{{Cite web|url=https://ethw.org/w/images/1/17/Dilhac.pdf|title = Engineering and Technology History Wiki}}</ref> From 1803 on, the French also used the 3-arm Depillon semaphore at coastal locations to provide warning of British incursions.<ref name="Burns2004"/>

====Chappe system technical operation====
The Chappe brothers determined by experiment that it was easier to see the angle of a rod than to see the presence or absence of a panel. Their semaphore was composed of two black movable wooden arms, connected by a cross bar; the positions of all three of these components together indicated an alphabetic letter. With counterweights (named ''forks'') on the arms, the Chappe system was controlled by only two handles and was mechanically simple and reasonably robust. Each of the two 2-metre-long arms could display seven positions, and the 4.6-metre-long cross bar connecting the two arms could display four different angles, for a total of 196 symbols (7×7×4). Night operation with lamps on the arms was unsuccessful.<ref>Holzmann & Pehrson, p. 213</ref> To speed up transmission and to provide some semblance of security, a [[Code|code]] book was developed for use with semaphore lines. The Chappes' corporation used a code that took 92 of the basic symbols two at a time to yield 8,464 coded words and phrases.

The revised Chappe system of 1795 provided not only a set of codes but also an operational protocol intended to maximize line [[Throughput|throughput]]. Symbols were transmitted in cycles of "2 steps and 3 movements."

* '''Step 1, movement 1 (setup)''': The operator turned the indicator arms to align with the cross bar, forming a non-symbol, and then turned the cross bar into position for the next symbol.
* '''Step 1, movement 2 (transmission)''': The operator positioned the indicator arms for current symbol and waited for the downline station to copy it.
* '''Step 2, movement 3 (completion)''': The operator turned the cross bar to a vertical or horizontal position, indicating the end of a cycle.

In this manner, each symbol could propagate down the line as quickly as operators could successfully copy it, with [[Acknowledgement (data networks)|acknowledgement]] and [[Flow control (data)|flow control]] built into the protocol. A symbol sent from Paris took 2 minutes to reach Lille through 22 stations and 9 minutes to reach Lyon through 50 stations.  A rate of 2–3 symbols per minute was typical, with the higher figure being prone to errors.  This corresponds to only 0.4–0.6 wpm, but with messages limited to those contained in the code book, this could be dramatically increased.<ref>Holzmann & Pehrson</ref><ref>{{cite web |url=http://chappe.ec-lyon.fr/message.html |title=Les dépêches |website=chappe.ec-lyon.fr |access-date=11 January 2022 |archive-url=https://web.archive.org/web/20140202183712/http://chappe.ec-lyon.fr/message.html |archive-date=2 February 2014 |url-status=dead}}</ref>

====History in France====
[[File:Reseau chappe77.png|thumb|right|The Chappe Network in France]]
After Chappe's initial line (between Paris and Lille), the Paris to Strasbourg with 50 stations followed soon after (1798). Napoleon Bonaparte made full use of the telegraph by obtaining speedy information on enemy movements. In 1801 he had Abraham Chappe build an extra-large station to transmit across the English Channel in preparation for an invasion of Britain.  A pair of such stations were built on a test line over a comparable distance.  The line to Calais was extended to Boulogne in anticipation and a new design station was briefly in operation at Boulogne, but the invasion never happened.  In 1812, Napoleon took up another design of Abraham Chappe for a mobile telegraph that could be taken with him on campaign.  This was still in use in 1853 during the Crimean War.<ref>Holzmann & Pehrson, pp. 71–73</ref>

The invention of the telegraph was followed by an enthusiasm concerning its potential to support [[Philosophy:Direct democracy|direct democracy]]. For instance, based on [[Biography:Jean-Jacques Rousseau|Rousseau]]'s argument that direct democracy was improbable in large constituencies, the French Intellectual [[Biography:Alexandre-Théophile Vandermonde|Alexandre-Théophile Vandermonde]] commented:

{{Blockquote|Something has been said about the telegraph which appears perfectly right to me and gives the right measure of its importance. Such invention might be enough to render democracy possible in its largest scale. Many respectable men, among them Jean-Jacques Rousseau, have thought that democracy was impossible within large constituencies.… The invention of the telegraph is a novelty that Rousseau did not expect to happen. It enables long-distance communication at the same pace and clarity than that of conversation in a living room. This solution may address by itself the objections to large [direct] democratic republics. It may even be done in the absence of representative constitutions.<ref>{{Cite journal|last=Mattelart|first=Armand|date=1999|title=La communication et la promesse de rédemption|url=https://www.persee.fr/doc/quad_0987-1381_1999_num_40_1_1428|journal=Quaderni|volume=40|issue=1|pages=69–78|doi=10.3406/quad.1999.1428}}</ref>}}

The operational costs of the telegraph in the year 1799/1800 were 434,000 francs ($1.2 million in 2015 in silver costs<ref name=Edvinsson>Rodney Edvinsson, [https://www.historicalstatistics.org/Currencyconverter.html Historical Currency Converter, accessed 8 January 2021.]</ref>).  In December 1800, Napoleon cut the budget of the telegraph system by 150,000 francs ($400,000 in 2015)<ref name=Edvinsson/> leading to the Paris-Lyons line being temporarily closed.  Chappe sought commercial uses of the system to make up the deficit, including use by industry, the financial sector, and newspapers.  Only one proposal was immediately approved—the transmission of results from the state-run lottery.  No non-government uses were approved.  The lottery had been abused for years by fraudsters who knew the results, selling tickets in provincial towns after the announcement in Paris, but before the news had reached those towns.<ref>{{multiref|Shelby T. McCloy, ''French Inventions of the Eighteenth Century'', p. 46, University Press of Kentucky, 2015 {{isbn|0813163978}}.|Rollo Appleyard, ''Pioneers of Electrical Communication'', pp. 271–272, Books for Libraries Press, 1968 (reprint of Macmillan, 1930) {{oclc|682063110}}.}}</ref>

[[File:Tour du telegraphe Chappe Saverne 02.JPG|thumb|left|A Chappe semaphore tower near Saverne, France]]
In 1819 Norwich Duff, a young British Naval officer, visiting Clermont-en-Argonne, walked up to the telegraph station there and engaged the signalman in conversation. Here is his note of the man's information:<ref>{{usurped|1=[https://web.archive.org/web/20070624023608/http://www.kittybrewster.com/images/Norwich_Duff_Journal_Intro.htm Commander Norwich Duff's European Tour Journal, 1819]}}, www.kittybrewster.com, archived 24 June 2007.</ref>

{{Blockquote|The pay is twenty five ''sous'' per day and he [the signalman] is obliged to be there from day light till dark, at present from half past three till half past eight; there are only two of them and for every minute a signal is left without being answered they pay five ''sous'': this is a part of the branch which communicates with Strasburg and a message arrives there from Paris in six minutes it is here in four.<ref>{{usurped|1=[https://web.archive.org/web/20050123042146/http://www.kittybrewster.com/images/Norwich_Duff_Journal_pages022.htm Journal of Norwich Duff]}}, 13 July 1819.</ref>|Norwich Duff}}

The network was reserved for government use, but an early case of wire fraud occurred in 1834 when two bankers, François and Joseph Blanc, bribed the operators at a station near [[Earth:Tours|Tours]] on the line between Paris and [[Earth:Bordeaux|Bordeaux]] to pass Paris stock exchange information to an accomplice in Bordeaux. It took three days for the information to travel the 300 mile distance, giving the schemers plenty of time to play the market.  An accomplice at Paris would know whether the market was going up or down days before the information arrived in Bordeaux via the newspapers, after which Bordeaux was sure to follow.  The message could not be inserted in the telegraph directly because it would have been detected.  Instead, pre-arranged deliberate errors were introduced into existing messages which were visible to an observer at Bordeaux.  Tours was chosen because it was a division station where messages were purged of errors by an inspector who was privy to the secret code used and unknown to the ordinary operators.  The scheme would not work if the errors were inserted prior to Tours.  The operators were told whether the market was going up or down by the colour of packages (either white or grey paper wrapping) sent by mail coach, or, according to another anecdote, if the wife of the Tours operator received a package of socks (down) or gloves (up) thus avoiding any evidence of misdeed being put in writing.<ref>{{cite book |last= Berloquin|first= Pierre|date= 2008|title= Hidden Codes & Grand Designs|publisher= Sterling|page= 25|isbn=978-1-4027-7300-6}}</ref>  The scheme operated for two years until it was discovered in 1836.<ref>Holzmann & Pehrson, pp. 75–76</ref><ref>{{cite web|last1=Holzmann |first1=Gerard J. |url=https://www.inc.com/magazine/19990915/13554.html |title=Taking stock |publisher=Inc |date=15 September 1999}}</ref>

The French optical system remained in use for many years after other countries had switched to the electrical telegraph.  Partly, this was due to inertia; France had the most extensive optical system and hence the most difficult to replace.  But there were also arguments put forward for the superiority of the optical system.  One of these was that the optical system is not so vulnerable to saboteurs as an electrical system with many miles of unguarded wire.  Samuel Morse failed to sell the electrical telegraph to the French government.  Eventually the advantages of the electrical telegraph of improved privacy, and all-weather and nighttime operation won out.<ref>{{cite web|last=Holzmann|first=Gerard|title=Data Communications: The First 2,500 Years|url=http://spinroot.com/gerard/pdf/hamburg94b.pdf|access-date=28 June 2011}}</ref>  A decision was made in 1846 to replace the optical telegraph with the Foy–Breguet electrical telegraph after a successful trial on the Rouen line.  This system had a display which mimicked the look of the Chappe telegraph indicators to make it familiar to telegraph operators.  Jules Guyot issued a dire warning of the consequences of what he considered to be a serious mistake.  It took almost a decade before the optical telegraph was completely decommissioned.  One of the last messages sent over the French semaphore was the report of the fall of Sebastopol in 1855.<ref>Holzmann & Pehrson, pp. 92–94</ref>

=== Sweden ===
[[File:Optical telegraph01 4484.jpg|thumb|A replica of an optical telegraph in Stockholm, Sweden]]
Sweden was the second country in the world, after France, to introduce an optical telegraph network.<ref name="Greene2016">David Greene, ''Light and Dark: An Exploration in Science, Nature, Art and Technology'', p. 159, CRC Press, 2016 {{isbn|1420034030}}.</ref>  Its network became the second most extensive after France.<ref>Holzmann & Pehrson, p. x</ref>  The central station of the network was at the Katarina Church in [[Place:Stockholm|Stockholm]].<ref>Edelcrantz, p. 174</ref>  The system was faster than the French system, partly due to the Swedish control panel<ref name="Holzmann & Pehrson, pp. 104–105">Holzmann & Pehrson, pp. 104–105</ref> and partly to the ease of transcribing the octal code (the French system was recorded as [[Pictogram|pictogram]]s).<ref name="Holzmann & Pehrson, p. 180">Holzmann & Pehrson, p. 180</ref>  The system was used primarily for reporting the arrival of ships, but was also useful in wartime for observing enemy movements and attacks.<ref name="Holzmann & Pehrson, p. 117">Holzmann & Pehrson, p. 117</ref>

The last stationary semaphore link in regular service was in Sweden, connecting an island with a mainland telegraph line. It went out of service in 1880.

====Development in Sweden<!-- "Sweden" required to avoid repeated heanding name per MOS:HEAD -->====
Inspired by news of the Chappe telegraph, the Swedish inventor Abraham Niclas Edelcrantz experimented with the optical telegraph in Sweden. He constructed a three-station experimental line in 1794 running from the royal castle in Stockholm, via Traneberg, to the grounds of Drottningholm Castle, a distance of {{convert|12|km}}.  The first demonstration was on 1 November, when Edelcrantz sent a poem dedicated to the king, Gustav IV Adolf, on his fourteenth birthday. On 7 November the king brought Edelcrantz into his Council of Advisers with a view to building a telegraph throughout Sweden, Denmark, and Finland.<ref>Holzmann & Pehrson, pp. 101–103</ref>

====Edelcrantz system technical operation====
After some initial experiments with Chappe-style indicator arms, Edelcrantz settled on a design with ten iron shutters.  Nine of these represented a 3-digit octal number and the tenth, when closed, meant the code number should be preceded by "A".  This gave 1,024 codepoints which were decoded to letters, words or phrases via a codebook.<ref>Holzmann & Pehrson, p. 103</ref>  The telegraph had a sophisticated control panel which allowed the next symbol to be prepared while waiting for the previous symbol to be repeated on the next station down the line.  The control panel was connected by strings to the shutters.  When ready to transmit, all the shutters were set at the same time with the press of a footpedal.<ref name="Holzmann & Pehrson, pp. 104–105"/>

The shutters were painted matte black to avoid reflection from sunlight and the frame and arms supporting the shutters were painted white or red for best contrast.<ref>Edelcrantz, p. 164</ref>  Around 1809 Edelcrantz introduced an updated design.  The frame around the shutters was dispensed with leaving a simpler, more visible, structure of just the arms with the indicator panels on the end of them.  The "A" shutter was reduced to the same size as the other shutters and offset to one side to indicate which side was the most significant digit (whether the codepoint is read left-to-right or right-to-left is different for the two adjacent stations depending on which side they are on).<ref name="Holzmann & Pehrson, p. 117"/>  This was previously indicated with a stationary indicator fixed to the side of the frame, but without a frame this was no longer possible.<ref>Edelcrantz, pp. 144, 146</ref>

The distance that a station could transmit depended on the size of the shutters and the power of the telescope being used to observe them.  The smallest object visible to the human eye is one that subtends an angle of 40 seconds of arc, but Edelcrantz used a figure of 4 minutes of arc to account for atmospheric disturbances and imperfections of the telescope.  On that basis, and with a 32× telescope, Edelcrantz specified shutter sizes ranging from 9 inches ({{nowrap|22 cm}}) for a distance of half a Swedish mile ({{nowrap|5.3 km}}) to 54 inches ({{nowrap|134 cm}}) for 3 Swedish miles ({{nowrap|32 km}}).<ref>Edelcrantz, pp. 166–167</ref>  These figures were for the original design with square shutters.  The open design of 1809 had long oblong shutters which Edelcrantz thought was more visible.<ref>Edelcrantz, p. 165</ref>  Distances much further than these would require impractically high towers to overcome the curvature of the Earth as well as large shutters.  Edelcrantz kept the distance between stations under 2 Swedish miles ({{nowrap|21 km}}) except where large bodies of water made it unavoidable.<ref>Edelcrantz, p. 169</ref>

The Swedish telegraph was capable of being used at night with lamps.  On smaller stations lamps were placed behind the shutters so that they became visible when the shutter was opened.  For larger stations, this was impractical.  Instead, a separate tin box matrix with glass windows was installed below the daytime shutters.  The lamps inside the tin box could be uncovered by pulling strings in the same way the daytime shutters were operated.  Windows on both sides of the box allowed the lamps to be seen by both the upstream and downstream adjacent stations.  The codepoints used at night were the complements of the codepoints used during the day.  This made the pattern of lamps in open shutters at night the same as the pattern of closed shutters in daytime.<ref>Edelcrantz, pp. 170–171</ref>

====First network: 1795–1809====
The first operational line, Stockholm to Vaxholm, went into service in January 1795.  By 1797 there were also lines from Stockholm to Fredriksborg, and Grisslehamn via Signilsskär to Eckerö in Åland.  A short line near Göteborg to Marstrand on the west coast was installed in 1799. During the War of the Second Coalition, Britain tried to enforce a blockade against France.  Concerned at the effect on their own trade, Sweden joined the Second League of Armed Neutrality in 1800.  Britain was expected to respond with an attack on one of the Nordic countries in the league.  To help guard against such an attack, the king ordered a telegraph link joining the systems of Sweden and Denmark.  This was the first international telegraph connection in the world.  Edelcrantz made this link between Helsingborg in Sweden and Helsingør in Denmark, across the Öresund, the narrow strait separating the two countries.  A new line along the coast from Kullaberg to Malmö, incorporating the Helsingborg link, was planned in support and to provide signalling points to the Swedish fleet.  Nelson's attack on the Danish fleet at [[Place:Copenhagen|Copenhagen]] in 1801 was reported over this link, but after Sweden failed to come to Denmark's aid it was not used again and only one station on the supporting line was ever built.<ref>Holzmann & Pehrson, pp. 105–109</ref>

In 1808 the Royal Telegraph Institution was created and Edelcrantz was made director.<ref>Holzmann & Pehrson, p. 114</ref>  The Telegraph Institution was put under the jurisdiction of the military, initially as part of the Royal Engineering Corps.<ref name="Holzmann & Pehrson, p. 120">Holzmann & Pehrson, p. 120</ref>  A new code was introduced to replace the 1796 codebook with 5,120 possible codepoints with many new messages.  The new codes included punishments for delinquent operators.  These included an order to the operator to stand on one of the telegraph arms (code 001-721), and a message asking an adjacent station to confirm that they could see him do it (code 001-723).<ref>Holzmann & Pehrson, p. 116</ref>  By 1809, the network had 50 stations over {{nowrap|200 km}} of line employing 172 people.<ref name="Holzmann & Pehrson, p. 117"/>  In comparison, the French system in 1823 had {{nowrap|650 km}} of line and employed over three thousand people.<ref name="Greene2016"/>

In 1808, the Finnish War broke out when Russia seized Finland, then part of Sweden.  Åland was attacked by Russia and the telegraph stations destroyed.  The Russians were expelled in a revolt, but attacked again in 1809.  The station at Signilsskär found itself behind enemy lines, but continued to signal the position of Russian troops to the retreating Swedes.  After Sweden ceded Finland in the Treaty of Fredrikshamn, the east coast telegraph stations were considered superfluous and put into storage.  In 1810, the plans for a south coast line were revived but were scrapped in 1811 due to financial considerations.  Also in 1811, a new line from Stockholm via Arholma to Söderarm lighthouse was proposed, but also never materialised.<ref>Holzmann & Pehrson, pp. 117–118</ref>  For a while, the telegraph network in Sweden was almost non-existent, with only four telegraphists employed by 1810.<ref>Holzmann & Pehrson, pp. 118, 120</ref>

====Rebuilding the network====
The post of Telegraph Inspector was created as early as 1811, but the telegraph in Sweden remained dormant until 1827 when new proposals were put forward.  In 1834, the Telegraph Institution was moved to the Topographical Corps.  The Corps head, Carl Fredrik Akrell, conducted comparisons of the Swedish shutter telegraph with more recent systems from other countries.  Of particular interest was the semaphore system of Charles Pasley in England which had been on trial in Karlskrona.  Tests were performed between Karlskrona and Drottningskär, and, in 1835, nighttime tests between Stockholm and Fredriksborg.  Akrell concluded that the shutter telegraph was faster and easier to use, and was again adopted for fixed stations.  However, Pasley's semaphore was cheaper and easier to construct, so was adopted for mobile stations.  By 1836 the Swedish telegraph network had been fully restored.<ref name="Holzmann & Pehrson, p. 120"/>

The network continued to expand.  In 1837, the line to Vaxholm was extended to Furusund.  In 1838 the Stockholm-Dalarö-Sandhamn line was extended to Landsort.  The last addition came in 1854 when the Furusund line was extended to Arholma and Söderarm.<ref name="Holzmann & Pehrson, p. 120"/>  The conversion to electrical telegraphy was slower and more difficult than in other countries.  The many stretches of open ocean needing to be crossed on the Swedish archipelagos was a major obstacle.  Akrell also raised similar concerns to those in France concerning potential sabotage and vandalism of electrical lines.  Akrell first proposed an experimental electrical telegraph line in 1852.  For many years the network consisted of a mix of optical and electrical lines.  The last optical stations were not taken out of service until 1881, the last in operation in Europe.  In some places, the [[Engineering:Heliograph|heliograph]] replaced the optical telegraph rather than the electrical telegraph.<ref>Holzmann & Pehrson, pp. 120–126</ref>}

===United Kingdom===
[[File:Murray Shutter Telegraph 1795.png|thumb|right|Diagram of UK Murray six-shutter system, with shutter 6 in the horizontal position, and shutters 1–5 vertical]]
In Ireland, Richard Lovell Edgeworth returned to his earlier work in 1794, and proposed a telegraph there to warn against an anticipated French invasion; however, the proposal was not implemented. Lord George Murray, stimulated by reports of the Chappe semaphore, proposed a system of visual telegraphy to the British Admiralty in 1795.<ref name="EdEnc1832"/> He employed rectangular framework towers with six five-foot-high octagonal shutters on horizontal axes that flipped between horizontal and vertical positions to signal.<ref name="Mech1828">[https://books.google.com/books?id=rNZQAAAAYAAJ&pg=PA298 Lieutenant Watson's Telegraph] Mechanics' magazine, Volume 8 No. 222, Knight and Lacey, 1828, pages 294-299</ref> The Rev. Mr Gamble also proposed two distinct five-element systems in 1795: one using five shutters, and one using five ten-foot poles.<ref name="EdEnc1832"/> The British Admiralty accepted Murray's system in September 1795, and the first system was the 15 site chain from London to Deal.<ref name="Wrixon2005">F.B. Wrixon (2005), {{ISBN|978-1-57912-485-4}} Codes, Ciphers, Secrets and Cryptic Communication pp. 444-445 cover Murray's shutter telegraph in the U.K., with codes.</ref> Messages passed from London to Deal in about sixty seconds, and sixty-five sites were in use by 1808.<ref name="Wrixon2005"/>
[[File:Stalbanshighstreet1807 cropped640.jpg|thumb|St. Albans High Street in 1807, showing the shutter telegraph on top of the city's Clock Tower. It was on the London to Great Yarmouth line.<ref name=OT>{{Cite web|url=http://www.vauxhallandkennington.org.uk/telegraph.shtml|title = Optical Telegraph}}</ref>]]
Chains of Murray's shutter telegraph stations were built along the following routes: [[Place:London|London]]–Deal and Sheerness, London–Great Yarmouth, and London–Portsmouth and Plymouth.<ref name=OT/> The line to Plymouth was not completed until 4 July 1806, and so could not be used to relay the news of Trafalgar.{{sfn|Burns|2004|p=49}} The shutter stations were temporary wooden huts, and at the conclusion of the Napoleonic wars they were no longer necessary, and were closed down by the Admiralty in March 1816.<ref name="MSSC">Military Signals from the South Coast, John Goodwin, 2000</ref>

Following the Battle of Trafalgar, the news was transmitted to London by frigate to Falmouth, from where the captain took the dispatches to London by coach along what became known as the Trafalgar Way; the journey took 38 hours. This delay prompted the Admiralty to investigate further.

A replacement semaphore system was sought, and of the many ideas and devices put forward the Admiralty chose the simpler semaphore system invented by Sir Home Popham.<ref name="EncyBrit1824"/><ref name="EdEnc1832"/> A Popham semaphore was a single fixed vertical 30 foot pole, with two movable 8 foot arms attached to the pole by horizontal pivots at their ends, one arm at the top of the pole, and the other arm at the middle of the pole.<ref name="Burns2004"/><ref name="EncyBrit1824"/> The signals of the Popham semaphore were found to be much more visible than those of the Murray shutter telegraph.<ref name="Burns2004"/> Popham's 2-arm semaphore was modelled after the 3-arm Depillon French semaphore.<ref name="Burns2004"/> An experimental semaphore line between the Admiralty and Chatham was installed in July 1816, and its success helped to confirm the choice.<ref name="MSSC"/>

Subsequently, the Admiralty decided to establish a permanent link to Portsmouth and built a chain of semaphore stations. Work started in December 1820<ref name="MSSC"/> with Popham's equipment replaced with another two-arm system invented by Charles Pasley. Each of the arms of Pasley's system could take on one of eight positions and it thus had more codepoints than Popham's.<ref>Holzmann & Pehrson, p. 196</ref> In good conditions messages were sent from London to Portsmouth in less than eight minutes.<ref>C. I. Hamilton, ''The Making of the Modern Admiralty: British Naval Policy-Making, 1805–1927'', p. 92, Cambridge University Press, 2011 {{ISBN|9781139496544}}.</ref> The line was operational from 1822 until 1847, when the railway and electric telegraph provided a better means of communication. The semaphore line did not use the same locations as the shutter chain, but followed almost the same route with 15 stations: Admiralty (London), Chelsea Royal Hospital, Putney Heath, Coombe Warren, Coopers Hill, Chatley Heath, Pewley Hill, Bannicle Hill, Haste Hill (Haslemere), Holder Hill, (Midhurst), Beacon Hill, Compton Down, Camp Down, Lumps Fort (Southsea), and Portsmouth Dockyard. The semaphore tower at Chatley Heath, which replaced the Netley Heath station of the shutter telegraph, is currently being restored by the [[Organization:Landmark Trust|Landmark Trust]] as self-catering holiday accommodation.<ref>{{cite web |title=Our plans for Semaphore Tower |url=https://www.landmarktrust.org.uk/news-and-events/Current-Restoration-Projects/semaphore-tower-appeal/our-plans/ |website=Landmark Trust |access-date=21 March 2020}}</ref> There will be public access on certain days when the restoration is complete.<ref>{{cite web |last1=Curley |first1=Rebecca |title=Chatley Tower restoration to make landmark rentable |url=https://www.yourlocalguardian.co.uk/news/18249227.chatley-tower-restoration-make-landmark-rentable/ |website=Sutton & Croydon Guardian |publisher=Newsquest Media Group Ltd |access-date=21 March 2020}}</ref>

The Board of the Port of Liverpool obtained a Private Act of Parliament to construct a chain of Popham optical semaphore stations from Liverpool to Holyhead in 1825.<ref>''Faster Than The Wind, The Liverpool to Holyhead Telegraph'', Frank Large, an avid publication {{ISBN|0-9521020-9-9}}</ref> The system was designed and part-owned by Barnard L. Watson, a reserve marine officer, and came into service in 1827. The line is possibly the only example of an optical telegraph built entirely for commercial purposes. It was used so that observers at Holyhead could report incoming ships to the Port of Liverpool and trading could begin in the cargo being carried before the ship docked. The line was kept in operation until 1860 when a railway line and associated electrical telegraph made it redundant.<ref>Holzmann & Pehrson, p. 197</ref><ref>Seija-Riitta Laakso, ''Across the Oceans: Development of Overseas Business Information Transmission 1815-1875'', BoD - Books on Demand, 2018 {{isbn|9517469047}}</ref>{{rp|181–183}} Many of the prominences on which the towers were built ('[[Engineering:Telegraph hill|telegraph hill]]s') are known as Telegraph Hill<!--intentional link to DAB page--> to this day.

====British empire====

=====Ireland=====
In Ireland R.L. Edgeworth was to develop an optical telegraph based on a triangle pointer, measuring up to 16 feet in height. Following several years promoting his system, he was to get admiralty approval and engaged in its construction during 1803–1804. The completed system ran from Dublin to Galway and was to act as a rapid warning system in case of French invasion of the west coast of Ireland. Despite its success in operation, the receding threat of French invasion was to see the system disestablished in 1804.<ref>Adrian James Kirwan, 'R.L. Edgeworth and Optical Telegraphy in Ireland, c. 1790-1805' in Proceedings of the Royal Irish Academy (2017). https://www.jstor.org/stable/10.3318/priac.2017.117.02?seq=1#page_scan_tab_contents</ref>

=====Canada=====
In Canada, Prince Edward, Duke of Kent established the first semaphore line in North America. In operation by 1800, it ran between the city of Halifax and the town of Annapolis in Nova Scotia, and across the Bay of Fundy to Saint John and Fredericton in New Brunswick. In addition to providing information on approaching ships, the Duke used the system to relay military commands, especially as they related to troop discipline. The Duke had envisioned the line reaching as far as the British garrison at Quebec City, but the many hills and coastal fog meant the towers needed to be placed relatively close together to ensure visibility. The labour needed to build and continually man so many stations taxed the already stretched-thin British military and there is doubt the New Brunswick line was ever in operation. With the exception of the towers around Halifax harbour, the system was abandoned shortly after the Duke's departure in August 1800.<ref>{{Citation|last=Raddall|first=Thomas H.|title=Warden of the North|publisher=McClelland and Stewart Limited|place=Toronto, Canada|year=1971|url=http://www.ourroots.ca/toc.aspx?id=6183&amp;qryID=8cbb8294-7361-4fdc-ac50-78627a229b5f}}</ref><ref>{{Citation|last=Rens|first=Jean-Guy|title=The invisible empire: A history of the telecommunications industry in Canada|publisher=McGill-Queen's University Press|place=Montreal, Canada|year=2001|isbn=9780773520523|url=https://books.google.com/books?id=8kUmt-Q7ySMC&pg=PA7 }}</ref>

=====Malta=====
[[File:Sudika Nadur Ta' Kenuna.jpg|thumb|left|Ta' Kenuna Tower, a semaphore tower in Nadur, Gozo, Malta, built by the British in 1848]]
The British military authorities began to consider installing a semaphore line in Malta in the early 1840s. Initially, it was planned that semaphore stations be established on the bell towers and domes of the island's churches, but the religious authorities rejected the proposal. Due to this, in 1848 new semaphore towers were constructed at Għargħur and Għaxaq on the main island, and another was built at Ta' Kenuna on Gozo. Further stations were established at the Governor's Palace, Selmun Palace and the Giordan Lighthouse. Each station was staffed by the Royal Engineers.<ref>{{cite web|title=Semaphore Tower|url=http://gharghur.gov.mt/semaphore-tower/|website=Għargħur Local Council|archive-url=https://web.archive.org/web/20160304094408/http://gharghur.gov.mt/semaphore-tower/|archive-date=4 March 2016}}</ref>

=====India=====
[[File:Semaphore Tower at Khatirbazar, Andul.jpg|thumb|upright|The Semaphore Tower at Khatirbazar, Andul in Howrah district of West Bengal]]

In India, semaphore towers were introduced in 1810. A series of towers were built between Fort William, [[Place:Kolkata|Kolkata]] to Chunar Fort near Varanasi.The towers in the plains were {{cvt|75 – 80|ft}} tall and those in the hills were {{cvt|40 – 50|ft}} tall, and were built at an interval of about {{cvt|13|km}}.<ref>{{cite news |last1=Singh |first1=Gurvinder |title=Towering messengers of a bygone era |url=https://www.thehindubusinessline.com/blink/know/towering-messengers-of-a-bygone-era/article23923909.ece |access-date=31 March 2019 |issue=Business Line |date=18 May 2018}}</ref>

=====Van Diemen's Land=====
In southern Van Diemens Land ( Tasmania ) a signalling system to announce the arrival of ships was suggested by Governor-In-Chief Lachlan Macquarie when he made his first visit in 1811 <ref>{{cite web |title=Maquarie's Journals - Saturday 30th. Novr. 1811.| url=https://www.mq.edu.au/macquarie-archive/journeys/1811/1811a/nov30.html|website=Journeys In Time }}</ref>  
Initially a simple flag system in 1818 between Mt.Nelson and Hobart, it developed into a system with two revolving arms by 1829, the system was quite crude and the arms were difficult to operate. 
In 1833 Charles O'Hara Booth took over command of the Port Arthur penal settlement, as an "enthusiast in the art of signalling" <ref>{{cite book |last1=Masters |first1=W.E. |title=The Semaphore Telegraph System of Van Diemen's Land |date=1973 |publisher=Cat & Fiddle Press |isbn=0-85853-009-0 |page=8}}</ref> he saw the value of better communications with the headquarters in Hobart. 
During his command the semaphore system was extended to include 19 stations on the various mountains and islands between Port Arthur and Hobart. Until 1837 three single rotating arm semaphores were used. 
Subsequently the  network was upgraded to use signal posts with six arms - a pair top, middle and bottom. 
This enabled the semaphore to send 999 signal codes. 
Captain George King of the Port Office and Booth together contributed to the code book for the system.<ref>{{cite book |last1=Masters |first1=W.E. |title=The Semaphore Telegraph System of Van Diemen's Land |date=1973 |publisher=Cat & Fiddle Press |isbn=0-85853-009-0 |page=14}}</ref> 
King drew up shipping related codes and Booth added Government, Military and penal station matters. 
In 1877 Port Arthur was closed and the semaphore was operated for shipping signals only, it was finally replaced with a simple flagstaff after the introduction of the telephone in 1880.

[[File:LowHeadSemaphore.jpg|thumb|left|alt=A restored two-arm semaphore post at Low Head in Tasmania|A restored two-arm semaphore post at Low Head in Tasmania]]
In the north of the state there was a requirement to report on shipping arrivals as they entered the Tamar Estuary, some 55 kilometers from the main port at this time in Launceston. The Tamar Valley Semaphore System was based on a design by Peter Archer Mulgrave.<ref>{{cite book|title=Peter Archer Mulgrave|chapter=Mulgrave, Peter Archer (1778–1847)|chapter-url=https://adb.anu.edu.au/biography/mulgrave-peter-archer-2489|website=Australian Dictionary of Biography|publisher=National Centre of Biography, Australian National University}}</ref> This design used two arms, one with a cross piece at the end. The arms were rotated by ropes, and later chains. The barred arm positions indicated numbers 1 to 6 clockwise from the bottom left and the unbarred arm 7,8,9, STOP and REPEAT.[[File:FanPositions2.png|thumb|right|alt=Fan positions for the Mulgrave design used in the Tamar Valley Semaphore System|The vane positions indicate code numbers]] A message was sent by sending numbers sequentially to make up a code. As with other systems the code was decoded via a code book. On 1 October 1835 it was announced in the Launceston Advertiser - "...that the signal stations are now complete from Launceston to George Town, and communication may he made, as well as received,from the Windmill Hill to George Town, in a very few minutes, on a clear day".<ref>{{cite web|title=Launceston Advertiser - 1 Oct 1835|url=https://trove.nla.gov.au/newspaper/article/84775296?searchTerm=semaphore# |newspaper=Launceston Advertiser|date = October 1835}}</ref> The system comprised six stations - Launceston Port Office, Windmill Hill, Mt. Direction, Mt.George, George Town Port Office, Low Head lighthouse. The Tamar Valley semaphore telegraph operated for twenty-two and a half years closing on 31 March 1858 after the introduction of the electric telegraph.<ref>{{cite book |last1=Shipp |first1=Wayne |title=The Tamar Valley Semaphore Telegraph |date=2014 |publisher=Low Head Pilot Station Museum |isbn=978-0-646-93206-4 |page=37}}</ref>
In the 1990s the Tamar Valley Signal Station Committee Inc. was formed to restore the system. The works were carried out over several years and the semaphore telegraph was declared complete once more on Sunday 30 September 2001.<ref>{{cite book |last1=Shipp |first1=Wayne |title=The Tamar Valley Semaphore Telegraph |date=2014 |publisher=Low Head Pilot Station Museum |isbn=978-0-646-93206-4 |page=ix}}</ref>

===Iberia===
[[File:Telégrafo óptico, Ávila.jpg|thumb|150px|Restored semaphore in Adanero, Spain.]]

==== Spain ====
In Spain, the engineer Agustín de Betancourt developed his own system which was adopted by that state; in 1798 he received a Royal Appointment,<ref>{{Cite web|date=2020-04-14|title=Agustín de Betancourt, uno de los ingenieros más prestigiosos de Europa|url=https://museopostalytelegrafico.es/agustin-de-betancourt/|access-date=2020-10-18|website=Museo Postal y Telegráfico|language=es}}</ref> and the first stretch of line connecting [[Place:Madrid|Madrid]] and Aranjuez was in operation as of August 1800.<ref>{{Cite news|date=2002-02-04|title=Nostalgia del telégrafo óptico|language=es|work=El País|url=https://elpais.com/diario/2002/02/05/madrid/1012911883_850215.html|access-date=2020-10-18|issn=1134-6582}}</ref> Spain was spanned by an extensive semaphore telegraph network in the 1840s and 1850s.<ref name="Roig">{{cite book |last1=Roig |first1=Sebastián Olivé |title=Historia de la telegrafía óptica en España |date=1990 |publisher=MINISTERIO DE TRANSPORTE, TURISMO Y COMUNICACIONES |location=Madrid |url=http://forohistorico.coit.es/index.php/biblioteca/libros-electronicos/item/historia-de-la-telegrafia-optica-en-espana |access-date=10 January 2019}}</ref> The three main semaphore lines radiated from [[Place:Madrid|Madrid]].<ref name="Roig" /><ref name="Telegrafos2014">{{cite book |last1=Fundación Telefónica |title=Telégrafos. Un relato de su travesía centenaria |date=2014 |publisher=Grupo Planeta Spain |isbn=978-8408129653 |url=https://books.google.com/books?id=ANkcAwAAQBAJ&pg=PA1843 |access-date=10 January 2019}}</ref> The first ran north to Irun on the Atlantic coast at the French border. The second ran east to the Mediterranean, then north along the coast through [[Place:Barcelona|Barcelona]] to the French border. The third ran south to Cadiz on the Atlantic coast. These lines served many other Spanish cities, including: Aranjuez, Badajoz, Burgos, Castellon, Ciudad Real, Córdoba, Cuenca, Gerona, Pamplona, San Sebastian, Seville, Tarancon, Taragona, [[Place:Toledo, Spain|Toledo]], Valladolid, [[Place:Valencia|Valencia]], Vitoria and Zaragoza.<ref name="Telegrafos2014" />

The rugged topography of the Iberian peninsula that facilitated the design of semaphore lines conveying information from hilltop to hilltop, made it difficult to implement wire telegraph lines when that technology was introduced in the mid 19th century. The Madrid-Cadiz line was the first to be dismantled in 1855, but other segments of the optical system continued to function until the end of the Carlist Wars in 1876.<ref>Aguilar Pérez, Antonio; y Martínez Lorente, Gaspar, La telegrafía óptica en Cataluña, Scripta Nova, revista electrónica de Geografía y Ciencias Sociales.Vol. VII, núm. 137, 15 de marzo de 2003. Universidad de Barcelona. ISSN 1138-9788.</ref>

==== Portugal ====
In Portugal, the British forces fighting Napoleon in Portugal soon found that the Portuguese Army had already a very capable semaphore terrestrial system working since 1808, giving the Duke of Wellington a decisive advantage in intelligence. The innovative Portuguese telegraphs, designed by {{Interlanguage link|Francisco António Ciera|pt}}, a mathematician, were of 3 types: 3 shutters, 3 balls and 1 pointer/moveable arm.<ref>{{Cite journal |last=Thompson |first=Mark S. |date=1 December 2016 |title=Allied Use of Telegraphs during the Peninsular War |url=https://www.napoleon-series.org/military-info/organization/Britain/Miscellaneous/Telegraphs.pdf |journal=Royal Engineers Journal |via=The Napoleon Series}}</ref> <!-- (the first for longer distances, the other two for short) and with the advantage of all having only 6 significant positions. -->He also wrote the code book "Táboas Telegráphicas"<!-- , with 1554 entries from 1 to 6666 (1 to 6, 11 to 16,... 61 to 66, 111 to 116,... etc.) -->, the same for the 3 telegraph types.  Since early 1810 the network was operated by "Corpo Telegráfico", the first Portuguese military Signal Corps.<ref>{{Cite journal |last1=Luna |first1=Isabel de |last2=Sousa |first2=Ana Catarina |last3=Leal |first3=Rui Sá |date=2008 |title=Telegrafia visual na Guerra Peninsular. 1807-1814 |url=https://repositorio.ul.pt/handle/10451/40114 |journal=Boletim Cultural da Câmara Municipal de Mafra |pages=67–136}}</ref>

===Other regions===
[[File:BremerhavenSemaphore.jpg|thumb|Optical telegraph in the harbour of Bremerhaven, Germany]]
Once it had proved its success in France, the optical telegraph was imitated in many other countries, especially after it was used by Napoleon to coordinate his empire and army. In most of these countries, the postal authorities operated the semaphore lines.  Many national services adopted signalling systems different from the Chappe system. For example, the UK and Sweden adopted systems of shuttered panels (in contradiction to the Chappe brothers' contention that angled rods are more visible).  In some cases, new systems were adopted because they were thought to be improvements.  But many countries pursued their own, often inferior, designs for reasons of national pride or not wanting to copy from rivals and enemies.<ref>Holzmann & Pehrson, pp. 179–180</ref>

In 1801, the Danish post office installed a semaphore line across the Great Belt strait, ''Storebæltstelegrafen'', between islands Funen and Zealand with stations at [[Place:Nyborg|Nyborg]] on Funen, on the small island Sprogø in the middle of the strait, and at Korsør on Zealand. It was in use until 1865.<ref>[http://www.ptt-museum.dk/en/exhibitions/the_permanent_exhibition/the_age_of_invention/ The Age of Invention 1849–1920] {{Webarchive|url=https://web.archive.org/web/20110719130848/http://www.ptt-museum.dk/en/exhibitions/the_permanent_exhibition/the_age_of_invention/ |date=2011-07-19 }}, Post & Tele Museum Danmark, website visited on 8 May 2010.</ref>

{{main|Prussian semaphore system}}[[File:Semaphore Line Tower on Winter Palace.jpg|thumb|Former optical telegraph tower on the Winter Palace in Saint Petersburg, Russia]]
In the [[History:Kingdom of Prussia|Kingdom of Prussia]], Frederick William III ordered the construction of an experimental line in 1819, but due to the procrastination of defence minister Karl von Hake, nothing happened until 1830 when a short three-station line between Berlin and [[Place:Potsdam|Potsdam]] was built.  The design was based on the Swedish telegraph with the number of shutters increased to twelve.<ref>Holzmann & Pehrson, pp. 184–185</ref>  ''Postrat'' Carl Pistor proposed instead a semaphore system based on Watson's design in England.  An operational line of this design running Berlin-[[Place:Magdeburg|Magdeburg]]-Dortmund-Köln-[[Place:Bonn|Bonn]]-Koblenz was completed in 1833.  The line employed about 200 people, comparable to Sweden, but no network ever developed and no more official lines were built.  The line was decommissioned in 1849 in favour of an electrical line.<ref>Holzmann & Pehrson, pp. 185–187</ref>

Although there were no more government sponsored official lines, there was some private enterprise.  Johann Ludwig Schmidt opened a commercial line from [[Place:Hamburg|Hamburg]] to Cuxhaven in 1837.  In 1847, Schmidt opened a second line from [[Place:Bremen|Bremen]] to Bremerhaven.  These lines were used for reporting the arrival of commercial ships.  The two lines were later linked with three additional stations to create possibly the only private telegraph network in the optical telegraph era.<ref>Holzmann & Pehrson, p. 186</ref>  The telegraph inspector for this network was Friedrich Clemens Gerke, who would later move to the Hamburg-Cuxhaven electrical telegraph line and develop what became the International Morse Code.<ref>Huurdeman, p. 76</ref>  The Hamburg line went out of use in 1850, and the Bremen line in 1852.<ref>Holzmann & Pehrson, p. 187</ref>

In {{wipe|Russia}}, Tsar Nicolas I inaugurated a line between [[Place:Moscow|Moscow]] and [[Place:Warsaw|Warsaw]] of {{convert|1200|km}} length in 1833; it needed 220 stations staffed by 1,320 operators.  The stations were noted to be unused and decaying in 1859, so the line was probably abandoned long before this.<ref name="Holzmann & Pehrson, p. 180"/>

In the {{wipe|United States}}, the first optical telegraph was built by Jonathan Grout in 1804 but ceased operation in 1807. This {{convert|104|km|adj=on}} line between Martha's Vineyard with Boston transmitted shipping news.  An optical telegraph system linking Philadelphia and the mouth of the Delaware Bay was in place by 1809 and had a similar purpose; a second line to {{wipe|New York City}} was operational by 1834, when its Philadelphia terminus was moved to the tower of the Merchants Exchange. One of the principal hills in {{wipe|San Francisco}}, {{wipe|California}} is also named "Telegraph Hill", after the semaphore telegraph which was established there in 1849 to signal the arrival of ships into San Francisco Bay.

==As first data networks==
The optical telegraphs put in place at the turn of the 18th/19th centuries were the first examples of data networks.<ref>Holzmann & Pehrson, p. 214</ref>  Chappe and Edelcrantz independently invented many features that are now commonplace in modern networks, but were then revolutionary and essential to the smooth running of the systems.  These features included control characters, [[Routing|routing]], error control, [[Flow control (data)|flow control]], [[Social:Message precedence|message priority]] and [[Symbol rate|symbol rate control]].  Edelcrantz documented the meaning and usage of all his control codes from the start in 1794.  The details of the early Chappe system are not known precisely; the first operating instructions to survive date to 1809 and the French system is not as fully explained as the Swedish.<ref>Holzmann & Pehrson, pp. 210–216</ref>

Some of the features of these systems are considered advanced in modern practice and have been recently reinvented.  An example of this is the error control codepoint 707 in the Edelcrantz code.  This was used to request the repeat of a specified recent symbol.  The 707 was followed by two symbols identifying the row and column in the current page of the logbook that it was required to repeat.  This is an example of a selective repeat and is more efficient than the simple go back n strategy used on many modern networks.<ref name="Holzmann & Pehrson, p. 216">Holzmann & Pehrson, p. 216</ref>  This was a later addition; both Edelcrantz (codepoint 272), and Chappe (codepoint 2H6)<ref group=note>The notation here follows that given in Holzmann & Pehrson (p. 211).  The two digits represent, respectively the angle of the left and right indicators.  Vertical pointing up is "1" and each successive 45° from this position increments this number.  "H" means the regulator is in the horizontal position and "V" the vertical.</ref> initially used only a simple "erase last character" for error control, taken directly from Hooke's 1684 proposal.<ref>Holzmann & Pehrson, pp. 214–215</ref>

Routing in the French system was almost permanently fixed; only Paris and the station at the remote end of a line were allowed to initiate a message.  The early Swedish system was more flexible, having the ability to set up message connections between arbitrary stations.  Similar to modern networks, the initialisation request contained the identification of the requesting and target station.  The request was acknowledged by the target station by sending the complement of the code received.  This protocol is unique with no modern equivalent.<ref name="Holzmann & Pehrson, p. 216"/>  This facility was removed from the codebook in the revision of 1808.  After this, only Stockholm would normally initiate messages with other stations waiting to be [[Polling (computer science)|polled]].<ref name="Holzmann & Pehrson, p. 216"/>

The Prussian system required the Coblenz station (at the end of the line) to send a "no news" message (or a real message if there was one pending) back to Berlin on the hour, every hour.  Intermediate stations could only pass messages by replacing the "no news" message with their traffic.  On arrival in Berlin, the "no news" message was returned to Coblenz with the same procedure.  This can be considered an early example of a [[Engineering:Token passing|token passing]] system.  This arrangement required accurate clock synchronisation at all the stations.  A synchronisation signal was sent out from Berlin for this purpose every three days.<ref>Holzmann & Pehrson, p. 188</ref>
 
Another feature that would be considered advanced in a modern electronic system is the dynamic changing of transmission rates.  Edelcrantz had codepoints for faster (770) and slower (077).  Chappe also had this feature.<ref>Holzmann & Pehrson, pp. 215–216</ref>

==In popular culture==
[[File:Télégraphe Chappe par Töpffer.jpg|thumb|A cartoon strip of "Monsieur Pencil" (1831) by Rodolphe Töpffer]]

By the mid-19th century, the optical telegraph was well known enough to be referenced in popular works without special explanation. The Chappe telegraph appeared in contemporary fiction and comic strips. In "Mister Pencil" (1831), a comic strip by Rodolphe Töpffer, a dog fallen on a Chappe telegraph's arm—and its master attempting to help get it down—provoke an international crisis by inadvertently transmitting disturbing messages. In ''Lucien Leuwen'' (1834), Stendhal pictures a power struggle between Lucien Leuwen and the prefect M. de Séranville with the telegraph's director M. Lamorte. In Chapter 60 ("The Telegraph") of Alexandre Dumas' ''The Count of Monte Cristo'' (1844), the title character describes with fascination the semaphore line's moving arms: "I had at times seen rise at the end of a road, on a hillock and in the bright light of the sun, these black folding arms looking like the legs of an immense beetle."<ref>[http://www.ebooksgratuits.com/pdf/dumas_monte_cristo_3.pdf Page 84 in LE COMTE DE MONTE-CRISTO Tome III]</ref> He later bribes a semaphore operator to relay a false message in order to manipulate the French financial market. Dumas also describes in detail the functioning of a Chappe telegraph line. In Hector Malot's novel ''Romain Kalbris'' (1869), one of the characters, a girl named Dielette, describes her home in Paris as "...next to a church near which there was a clock tower. On top of the tower there were two large black arms, moving all day this way and that. [I was told later] that this was Saint-Eustache church and that these large black arms were a telegraph."<ref>[http://www.inlibroveritas.net/lire/oeuvre23495.html#page_163 See second paragraph in]</ref>

In the 21st century, the optical telegraph concept is mainly kept alive in popular culture through fiction such as the novel ''Pavane'' and  Terry Pratchett's "Clacks" in his Discworld novels,<ref>{{cite book |last1=Webb |first1=Simon |title=The Real World of Victorian Steampunk: Steam Planes & Radiophones |date=2019 |publisher=Pen and Sword |isbn=978-1526732866 |pages=49–51 |url=https://books.google.com/books?id=Ak8IEAAAQBAJ&pg=PT49}}</ref> most notably the 2004 novel ''Going Postal''.<ref>{{cite web |last1=Hague |first1=Jim |title=The Clacks in Discworld and Roundworld |url=https://accu.org/video/spring-2018-day-3/hague/ |website=ACCU.org |publisher=ACCU |access-date=28 December 2022}}</ref>

== See also ==
* [[Social:History of telecommunication|History of telecommunication]]
* Telegraph code, for more information on many of the codes used
* [[Engineering:Optical communication|Optical communication]]
* [[Engineering:Polybius square|Polybius square]]
* Railway signalling
* San Jose Semaphore
* [[Semaphore Flag Signaling System]]
* [[Social:Signal lamp|Signal lamp]]
* Telegraph Hill, for a list of telegraph hills

== Notes ==
{{Reflist|group=note}}

== References ==
; Citations
{{Reflist}}

; Bibliography
* {{cite book |url=https://books.google.com/books?id=7eUUy8-VvwoC |isbn=978-0-86341-327-8 |title=Communications: an international history of the formative years |first=R. W. |last=Burns |year=2004 }}
* Crowley, David and Heyer, Paul (ed) (2003) 'Chapter 17: The optical telegraph' ''Communication in History: Technology, Culture and Society'' (Fourth Edition) Allyn and Bacon, Boston pp.&nbsp;123–125
* Edelcrantz, Abraham Niclas, ''Afhandling om Telegrapher'' ("A Treatise on Telegraphs"), 1796, as translated in ch. 4 of Holzmann & Pehrson.
* Holzmann, Gerard J.; Pehrson, Bjorn, ''The Early History of Data Networks'', John Wiley & Sons, 1995 {{isbn|0818667826}}.
* Huurdeman, Anton A., ''The Worldwide History of Telecommunications'', John Wiley & Sons, 2003 {{ISBN|0471205052}}.

== Further reading ==
* ''The Victorian Internet'', Tom Standage, Walker & Company, 1998, {{ISBN|0-8027-1342-4}}
* ''The Old Telegraphs'', Geoffrey Wilson, Phillimore & Co Ltd 1976 {{ISBN|0-900592-79-6}}
* ''Faster Than The Wind, The Liverpool to Holyhead Telegraph'', Frank Large, an avid publication {{ISBN|0-9521020-9-9}}
* ''The early history of data networks'', Gerard Holzmann and Bjorn Pehrson, Wiley Publ., 2003, {{ISBN|0-8186-6782-6}}
* {{cite book |last=Burns |first=R.W. |title=Communications: an international history of the formative years |date=2004 |chapter= Semaphore Signaling, Chapter 2 |isbn=978-0-86341-327-8 |publisher=Institution of Electrical Engineers |ref=none}}

== External links ==
* [https://web.archive.org/web/20030212061645/http://people.deas.harvard.edu/~jones/cscie129/images/history/chappe.html Chappe's semaphore] (an illustrated history of optical telegraphy)
* [http://www.vauxhallandkennington.org.uk/telegraph.shtml Webpage including a map of England's telegraph chains]
* [http://www.portsdown-tunnels.org.uk/ancient_sites/telegraph_p1.html Diagrams and maps of Murray's U.K. semaphore stations]
* [https://s3.amazonaws.com/files.qrz.com/t/gb2hst/SHUTTER_CHART.jpg Chart of Murray's shutter-semaphore code]
* [http://www.portsdown-tunnels.org.uk/ancient_sites/telegraph_p2.html Photo and diagrams of Popham's U.K. semaphore stations]
* [http://www.morsekey.net/duesicilie.html Map of visual telegraph (semaphore) and electrical telegraph lines in Italy, 1860] {{in lang|it}}
* [https://www.inc.com/magazine/19990915/13554.html Details on the history of the Blanc brothers fraudulant use of the Semophore line]
* [https://rfmoz.github.io/telop/index.html Live recreation of the Spanish optical telegraph code] {{in lang|es}}

{{Optical communication}}
{{Telecommunications}}

{{DEFAULTSORT:Semaphore Line}}
[[Category:History of telecommunications]]
[[Category:Optical communications]]
[[Category:Semaphore]]

{{Sourceattribution|Optical telegraph}}