Dr A. D. Morrison-Low is Principal Curator, History of Science and Photography, at National Museums Scotland, Edinburgh. Her Paper below on the development of early lighthouse optics first appeared in: Bart Grob and Hans Hooijmaijers (eds.), Who needs scientific instruments: conference on Scientific Instruments and their users (Leiden, 2006), pp89-97.

'It was a Dark and Stormy Night':

Instrument Makers and the Northern Lights

A. D. Morrison-Low, National Museums Scotland

Allen Simpson, a former colleague, published a paper in 1994 entitled 'François Soleil, Andrew Ross and William Cookson: the Fresnel Lens Applied'. In it, he showed how burning lenses evolved into the stepped echelon optical configuration of the lighthouse lens, and commented that, although the burning lens might be 'considered part of the conventional field of scientific instruments, lighthouse lenses probably would not, even though the only aspect that differentiates them is the area of application.'[1]

In the beginning, of course, there were no lights. Scotland‘s coasts were dark and dangerous, especially on the sparsely populated and windswept Atlantic western shores.[2] However, from early medieval times most external trade was eastwards across the North Sea, and small craft made their way across these relatively calm waters to the greater ports of Europe, in the Low Countries and the Baltic. But there were severe natural hazards to be found all the way up the east coast of Scotland, particularly at the mouths of the various estuaries around which most of the east coast royal burghs were located. The royal burghs were the early settlements that had been granted special privileges by the king, amongst which was the ability to trade outside the kingdom.[3]

Charting these dangerous coasts remained an inexact science until the late 17th century. Scotland had of course appeared in Ptolemy's Geographia, and appeared approximately in various medieval portolan charts, and seamen's 'rutters','before being reasonably well outlined by Gerard Mercator in 1564.[4] During the 16th century, cartographic knowledge was regarded as the prerogative of princes, and it was not until about 1600, with the flowering of Renaissance thought in Northern Europe, that this information came to be seen as publicly available, and especially important for the ordinary navigator.[5]

The first printed sea-atlas, printed in Leiden in 1584 by Lucas Jansz Waghenaer, the Spieghel der Zeevaerdt, included a chart of the east coast from Bamburgh to Aberdeen. But it was not until 1681 that a survey, undertaken by Captain Greenvile Collins, sponsored by the Admiralty and Charles II, that the British coasts – including the east coast of Scotland, and Orkney and Shetland - were made available in a publication entitled Great Britain's Coasting Pilot. The Scotsman John Adair, who was surveying the east coast at the same time, did not manage to publish his charts until 1703.[6]

Both Collins's and Adair's charts show the dangers of the eastern approaches, and also illustrate Scotland's earliest lighthouses. Both Aberdeen and Leith (Edinburgh's port on the Firth of Forth – 'firth' being the Scots word for estuary) had beacons – braziers kept alight at night - at their harbour entrances from at least the 15th century.[7] That at Leith was illustrated by Greenvile Collins, and was a welcoming sight to mariners who had navigated up the Forth.[8] But it was the unlit hazards that the increasing numbers of seafarers wished to anticipate.

The first major obstacle in the Firth of Forth – upon which Leith lies, gateway to the capital – is a rocky island named the Isle of May. It is about five and a half miles south-east of Anstruther, and about eleven miles north-west of North Berwick: its long axis runs from north-west to south-east, it is about a mile in length and at its widest is about a third of a mile.[9] It is characterised in a recent architectural guide as a 'Bare island in the Firth of Forth, the home of monks in the Middle Ages, of lighthouse keepers since the 17th century'.[10] Some local Fife landowners grouped together to petition the Scots government for the provision of a light, and a simple grate on a tower was set up in 1636. This was a two-storey building, with the grate on the top; after 1786 there was a rope and pulley windlass which was used to haul up the pans of coal.[11]

By this time, the Northern Lighthouse Board had been formed, and had placed its first four lights strategically on the sea routes around Scotland's stormy and dangerous northern and western coasts: at Kinnaird Head in 1787; at the Mull of Kintyre, at North Ronaldsay in Orkney, and at Eilean Glas on Scalpay on the Minch (the narrow sea passage between the Hebridean islands and the mainland) in 1788. This government board came into being after the Union of the Parliaments in 1707, and had – and still has – responsibility for the lighthouses of Scotland and the Isle of Man. In England, the Brethren of Trinity House has had that responsibility since at least the 13th century; and in Ireland, the Commissioners of Irish Lights was formed from previous authorities in 1867.[12]

And now to the influence of scientific instruments: clearly these were not required when lighthouses used coal-fired grates of the type encountered on the Isle of May. Neither were instruments in any way influential in the lighting of one of the more revolutionary design for lighthouses, that of the fourth tower to stand on the dangerous Eddystone reef in the English Channel, built by John Smeaton in 1756. Despite Smeaton serving an apprenticeship as an instrument maker in London, his form of lighting was no more dramatic than a chandelier fitted with twenty-four candles – which it has been calculated was visible at about five miles.[13]

By the time the Northern Lighthouse Board had begun to build their first four defining towers around the Scottish coasts, they were employing Thomas Smith, a lighting engineer, to assist them in their task. Smith, born into a seafaring family across the Firth of Tay from the important port of Dundee in 1752, had lost his father by shipwreck when merely a child. He was then apprenticed to a Dundee metal-worker, and emigrated to Edinburgh just as that city enlarged itself in 1770. His oil lamp business expanded and by 1787 he had the contract to provide street lighting for the Old and subsequently the New Towns of Edinburgh. By 1807, he had expanded this to light the streets of Perth, Stirling, Ayr, Haddington and Aberdeen. The Northern Lighthouse Board appointed Smith as their Engineer in January 1787, six months after his proposal that reflector lamps be substituted for the coal light on the Isle of May was rejected by Edinburgh's Chamber of Commerce.[14]

Smith was not the Northern Lighthouse Board's first choice: they had first invited Ezekiel Walker of King's Lynn, the designer of improved lighting at Hunstanton Lighthouse on the Norfolk coast, to put up their first four lights. Instead, after Walker had declined this invitation, Smith was sent to Norfolk to learn about the particular problems involved in lighthouse illumination.[15] For the next twenty years, starting with the conversion of Kinnaird Castle at Fraserburgh into a lighthouse, Thomas Smith was responsible for the illumination of thirteen lighthouses, mostly at remote locations.[16]

Smith improved the intensity of illumination by reflection (or the 'catoptric' system). He developed and made arrays of parabolic reflector oil lamps of his own design. These had a light source at their focus, and a curved reflector formed of small pieces of mirror set in plaster, which focused the light into a parallel beam which could be rotated in a horizontal plane.[17] Nonetheless, these proved to be much less efficient than the silvered-copper reflectors already in use in France, and which were soon installed elsewhere with Argand lamps with glass chimneys.[18]

After 1801, Smith and his son-in-law Robert Stevenson (who was to succeed him in business and at the Northern Lighthouse Board) began to make and install Argand lamps and silvered-copper reflectors, starting at another lighthouse on an island in the Firth of Forth, at Inchkeith in 1804. However, the glass facet reflectors continued to be used by the Northern Lighthouse Board in lights elsewhere until the 1820s.[19]

The Argand lamp was the invention of Aimé Argand, a Swiss in Geneva, who developed in the 1780s a relatively smokeless oil lamp which gave a steady flame and a more intense light than previously achieved. His lamp had two vertical concentric tubes of thin brass, which held between them a cylindrical wick. Air passed through the tube, ensuring that both sides of the lighted wick had an even temperature and good combustion. A glass chimney channeled the airflow to give further brilliance, and the wick's height could also be finely adjusted. By about 1815 this lamp could produce the same amount of light as seven or eight candles. Initially, this increased the demand for sperm oil – the oil from the sperm whale – to such an extent that the French resorted to colza oil, which uses the seed of a wild cabbage. By 1845, the British had turned to this, too.[20]

Meantime, Smith had retired from lighthouse illumination and construction, handing over this side of the business to his step-son and son-on-law Robert Stevenson in 1797, when the younger man was aged twenty-five; Robert Stevenson formally succeeded him as Engineer to the Northern Lighthouse Board in July 1808. He had attended natural philosophy classes of both Professor John Anderson at Glasgow University and Professor John Robison at Edinburgh University, although he did not take a degree. He had also served practical apprenticeships in metalwork and lamp installation, and throughout his life he maintained an acquaintance with Edinburgh's intelligentsia, and belonged to many of the scientific societies which flourished at that point.[21]

In an attempt to increase the intensity of the beam, spherical reflectors had been introduced in some French lighthouses in the 1780s and 1790s, but these had proved problematic. At Cordouan, near Bordeaux, the famous and ancient lighthouse had been the scene of a series of practical investigations and tests, involving French philosophers, clockmakers and scientists. Both the scientist Jean-Charles Borda and the instrument maker Etienne Lenoir were concerned, producing a rotating apparatus with spherical reflectors; however, mariners found these inadequate. When Léonor Fresnel replaced them with a revolving dioptric apparatus in 1823 (using lenses instead of reflectors), he concluded that the original reflectors had been inadequately set up.[22]

Robert Stevenson developed his own metal parabolic reflector in about 1810. To ensure accuracy, Stevenson had the reflector curves drawn by Professor John Leslie and their moulds made by the Edinburgh optician Alexander Adie. This type of optical apparatus was supplied by various brass founders, including James Milne of Edinburgh, to many British and colonial lighthouses between 1810 and 1880.[23] One set of revolving parabolic reflectors was installed in 1811 in Robert Stevenson's most important lighthouse, the Bell Rock, and was removed in 1843 to Newfoundland, where it continued in service until after 1872.[24] Apparently the Scottish reflectors were plated so thickly that they could afford the loss of silver through cleaning, and thus gave long service. The metallic silver was recovered by collecting and burning the cleaning cloths.[25] They were also set up using a gauge, which ensured that the lamp – and thus the beam - was always set up correctly after cleaning.[26] It is unclear whether gauges of this type have survived. However, one survival (initially in the Stevenson family, and now in the National Museums of Scotland) is a device which can set a series of reflectors on a frame for a fixed light (not a revolving light, such as was that at the Bell Rock). "In fixed lights", wrote Alan Stevenson, "it is necessary, in order to approach as near as possible to an equal distribution of the light over the horizon, to place the reflectors, with their axes somewhat less than that of the divergence of the reflected cone".[27] The instrument is signed by Adie & Son, and was made for the Stevenson firm.[28]

It is inappropriate to linger here over the building of the Bell Rock lighthouse and the engineering problems that were solved there by Robert Stevenson; instead, we shall examine the next form of illumination: through prisms and glass. Allen Simpson's paper, referred to earlier, discussed in some detail the beginnings of this technology: how Robert Stevenson acquired two square-framed panels from the Parisian optician François Soleil of the type being used at Cordouan at about the same time that a 'burning glass' was ordered from the same maker by the University of Edinburgh. A few years later, Fresnel developed a smaller cylindrical lens, also constructed by Soleil, as a harbour light. Robert Stevenson and the Northern Lighthouse Board were greatly impressed by the results they obtained from these lenses; however, the entire affair was muddied by a bitter argument of priority of invention over the Fresnel lens, by Sir David Brewster, a scientific figure who commanded respect across Europe.[29]

Early in 1831 a committee of the Northern Lighthouse Board was formed to compare the merits of lenses and reflectors, following a report by Stevenson on the lenses. Two years later, there were some practical experiments, which took place on three consecutive nights in February 1833. Under the direction of young Alan Stevenson and his father Robert, light was shone through lenses and by reflectors from temporary buildings put up near the coast on Gullane Hill, about twelve miles from Edinburgh, and their effects were observed from Calton Hill in central Edinburgh by some Fellows of the Royal Society of Edinburgh, including Sir David Brewster, as well as the lighthouse committee.[30]

Everyone present was impressed by the results. Robert Stevenson found that light from the lens was equal to seven to nine reflectors; but Brewster went further and demanded that all Scottish lighthouses should immediately be furnished with lenses. Stevenson argued for caution, as many of Brewster's proposals were extremely expensive – not least the crippling Excise dues on the optical glass. But Brewster reacted badly, publishing anonymously a savage article in the Edinburgh Review (a widely-read political journal), and this led almost immediately to a major public parliamentary enquiry.[31]

This enquiry resulted in a number of recommendations, mostly to do with organisation and finance, which need not detain us here. Of more significance to the technology, Robert Stevenson's eldest son Alan was sent off immediately to France by the Commissioners to inspect the French lights, and to meet Léonor Fresnel, successor to his famous brother as secretary of the Commission des Phares.[32] He also spent time in the workshops of the instrument-maker François Soleil, the clockmaker Pierre Basile Lepaute, and the lamp-maker Bordier Marcet, pupil and successor of Argand.

As Alan Stevenson was to write himself:"Although many experiments were made with lenses during the winter of every succeeding year [to 1825], it was not until the spring of 1834 that the Commissioners took decisive steps for deciding the question as to the comparative merits of the catoptric and dioptric system, by sending me, in the spring of 1834, on a mission to Paris, with full power to take such steps for acquiring a perfect knowledge of the dioptric system, and forming an opinion on its merits, as I should find necessary. The singular liberality with which I was received by M. Léonor Fresnel, brother of the late illustrious inventor of the system, and his successor as the Secretary of the Lighthouse Commission of France, afforded me the means of making such a report on my return, as induced the Commissioners to authorise me to remove the reflecting apparatus of the revolving light at Inchkeith, and substitute dioptric instruments in its place. This change was completed, and the light exhibited on the evening of 1st October 1835; and so great was the satisfaction which the change produced, that the Commissioners immediately instructed me to make a similar change at the fixed light of the Isle of May, where the new light was exhibited on the 22nd September 1836."[33]

And as R.W. Munro states: "The Commissioners now became the pioneers of the dioptric system in the British Isles".[34] Both Inchkeith in 1835 and the Isle of May the following year had improved refractors made by Messrs Cookson of Newcastle, business associates of Robert Stevenson, with the effect of tripling the brightness of the lights. Trinity House (the English lighthouse board) soon employed Alan Stevenson to direct the construction of their first revolving dioptric light for Start Point in Devon in 1836.[35] But his greatest challenge was to be the building of a rock lighthouse at Skerryvore, an Atlantic-lashed jagged rock set among lesser reefs 12 miles west of Tiree. It lies between the Hebrides and the north Irish coast, near a new shipping highway opening westwards towards North America.[36]

Alan Stevenson's account of Skerryvore, which is another epic comparable with his father's at the Bell Rock, discusses many of the problems he encountered and solved, but here we will focus on one particular and crucial issue: the correct setting of the Fresnel lenses and the accurate centring of the lamp itself within the lenses. In his account he gives full credit to the maker of the optics, which was the Parisian workshop of François Soleil; and to the maker of the machinery, including the lampwork, by the Edinburgh brass founder, John Milne.[37] Having learned much from his French visit, the Oxford Dictionary of National Biography states that:
"Skerryvore's revolving dioptric apparatus was the most advanced in the world at that time, with prismatic rings instead of mirrors below the central belt, thus extending the improved dioptric effect. Stevenson further improved its efficiency by introducing inclined astragals into the lantern. His improvements to the dioptric system, which included the conversion of Fresnel's narrow lenses in fixed systems into a truly cylindrical drum, led to its wider adoption."[38]

But the task of assembling the delicate but very heavy component lens panels into a complete rotating optic was complex. The site, high in the lighthouse lantern, was very cramped and comparatively inaccessible; and not only did the tall optic have to remain in balance during construction; its shape also had to be adjusted within narrow margins to ensure that the optimum optical performance was achieved. Alan Stevenson's diagrams show how the problem was resolved – using specially designed instruments, engineered for Stevenson by John Adie, to align the panels as tension was taken up at their boundaries.

Stevenson's diagrams demonstrate, as do the surviving instruments which correspond to them, that his ideas have been elegantly solved by John Adie. The first instrument (Figure 1) is to ensure that the lenses meet at 'the proper horizontal angle, so that their axes shall meet with the proper inclination in the focus'.[39] The second (Figure 2) ascertains the verticality of the main lenses, or sets the subsidiary lenses or mirrors at the required angle, with a spirit level; so this is in fact a special clinometer.[40] The third, and final example (Figure 3), is a device which tests the true position of the lamp itself in the context of the lenses.[41] It acts as a radius, with F being the central burner, 'while its point A touches the centres of the lenses. At B is a graduated slide [now seized], which allows the radius arm to be lengthened or shortened to suit various different focal distances.' The dotted outline shows the device being moved on to an adjacent lens.[42] Without these scientific instruments, constructed for just this specific purpose, the complex lenses of Skerryvore could not have shone out with clarity over the Atlantic waves.

Figure 1
Instrument to ensure Fresnel lenses are correctly focused, Adie & Son, Edinburgh, c.1840 (NMS T.1958.49). Designed by Alan Stevenson for use at Skerryvore, 1844.

Figure 2
Instrument for ensuring that the vertical lenses or mirrors are set at the correct angle, by Adie & Son, Edinburgh, c.1840 (NMS T.1958.48). Designed by Alan Stevenson for use at Skerryvore, 1844.

Figure 3
Instrument to test the true position of the lamp in the context of Fresnel lenses (NMS T.1958.56).Designed by Alan Stevenson for use at Skerryvore, 1844.

As it says in a maddening piece of doggerel that all British schoolchildren can recite: 'It was a dark and stormy night, and the captain said to the mate, "Tell me a story"; so the mate began. It was a dark and stormy night . . .', and so on, ad infinitum.[43] In the paper mentioned at the beginning of this piece, Allen Simpson wrote:
"The fact that precision apparatus is an integral part of many types of discipline emphasises the necessity for instrument historians to justify the relevance of the specialism to wider audiences. And it is often in the literature and archives of other disciplines (in this case civil engineering) that valuable evidence can be found that illuminates the capabilities and limitations of the industrial concerns that are central to the world of precision instruments."[44]

The allotted time is over, and, as ever, the exercise has raised more questions than it has answered. Although the night was dark and stormy, the links between the instrument maker and the Northern Lights was clearly a strong one, and together they illuminated the seas. In due course, I hope to find out more.

References and acknowledgments

My thanks to Chris Henry, Director of the Museum of Scottish Lighthouses, Fraserburgh, for access to trade literature obtained from the Northern Lighthouse Board; to my former colleague, Dr Allen Simpson, for his help and advice; and to the Library of the National Museums of Scotland for their unfailing swiftness in supplying me with books; and the photographers for their help in providing me with suitable images.

[1] A.D.C. Simpson, 'François Soleil, Andrew Ross and William Cookson: the Fresnel Lens Applied', Bulletin of the Scientific Instrument Society, No. 41 (1994), 16-19; quotation on p.18.

[2] Much of the information given in the next few paragraphs is based upon a paper written by A.D. Morrison-Low and A.D.C. Simpson, ' “For Those in Peril on the Sea”: Scottish Sea-Marking to 1787', in Jan Parmentier (ed.), Navigating the Northern Seas (forthcoming).

[3] Derek Hall, Burgess, Merchant and Priest: Burgh Life in the Scottish Medieval Town (Edinburgh, 2002), 20-26.

[4] D.G. Moir (ed.), The Early Maps of Scotland to 1850 third edition, 2 vols. (Edinburgh, 1983), I, 3-16.

[5] P.D.A. Harvey, Maps in Tudor England (London, 1993), 7; see also David Buisseret, The Mapmakers' Quest: Depicting New Worlds in Renaissance Europe (Oxford, 2003).

[6] Moir, op. cit. (4), II, 1-2.

[7] Sir James Marwick (ed.), Extracts from the Records of the Burgh of Edinburgh, 1528-1557 (Edinburgh, 1871), 275-6, 289, 295, quoted by R.W. Munro, Scottish Lighthouses (Stornoway, 1979), 22-23; John Stuart (ed.), Extracts from the Council Register of the Burgh of Aberdeen, 1398-1570 (Aberdeen, 1844), 361-2, also quoted by Munro, op. cit. 23.

[8] Greenvile Collins, Great Britain's Coasting Pilot (London, 1776), 23. Collins's book was published in eleven editions between 1693 and 1792.

[9] W.J. Eggeling, The Isle of May (Edinburgh and London, 1960), 5.

[10] John Gifford, The Buildings of Scotland: Fife (London, 1988), 320.

[11] Eggeling, op. cit. (9), 33-34.

[12] For general accounts of each of these boards and their work, see D.B. Hague and R. Christie, Lighthouses: their Architecture, History and Archaeology (Llandysul, 1975); K. Sutton-Jones, Pharos (Salisbury, 1985); D. Alan Stevenson, The World's Lighthouses before 1820 (Oxford, 1959). The most authoritative account of the work of the Northern Lighthouse Board is Munro, op. cit. (7). See also, Martin Boyle, Lighthouses: Four Countries – One Aim (Southampton, 1996).

[13] John Smeaton, A Narrative of the Building and Description of the Construction of the Edystone Lighthouse with Stone … (London, 1791).

[14] Jean Leslie and Roland Paxton, Bright Lights: The Stevenson Engineers 1752-1971(Edinburgh, 1999), 13-23.

[15] Stevenson, op. cit. (12), 291-2.

[16] For the history of the lighthouse at Kinnaird Head see Sarah Swallow, Lighthouses of Scotland: Kinnaird Head (Southampton, 1998); Thomas Smith's 13 lights (and those of the individual Stevensons) can be found listed in Appendix 2 of Leslie and Paxton, op. cit. (14), 191-5.

[17] [Thomas Smith], 'Reflectors for light-houses', Encyclopaedia Britannica, fifth edition (Edinburgh, 1817), XVII, 671; however, for a discussion about the claim whether this was invented by Smith, or by Walker, see Leslie and Paxton, op. cit. (14), 23.

[18] Stevenson, op. cit. (12), 61-71; 288-90.

[19] Leslie and Paxton, op. cit. (14), 22-23; Robert Stevenson, 'Inchkeith', Edinburgh Encyclopaedia 22 vols. (Edinburgh, 1808-30), XII, 9-12; and 'Lighthouse', ibid., XIII, 1-18.

[20] Stevenson, op. cit. (12), 61-62. For a comprehensive history, see Michael Schrøder, The Argand Burner: its origin and Development in France and England 1780-1800 (Odense, 1969).

[21] Leslie and Paxton, op. cit. (14), 24-34.

[22] Stevenson, op. cit. (12), 187-200.

[23] Stevenson, op. cit. (12), 294-5.

[24] Stevenson, op. cit. (12), 224.

[25] Stevenson, op. cit. (12), 284.

[26] Alan Stevenson, Account of the Skerryvore Lighthouse, with Notes on the Illumination of Lighthouses (Edinburgh, 1848), 226.

[27] Stevenson, op. cit. (26), 230.

[28] NMS.T.1958.50.

[29] Simpson, op. cit. (1); for Brewster, see A.D. Morrison-Low and J.R.R. Christie (eds.), 'Martyr of Science': Sir David Brewster, 1781-1868 (Edinburgh, 1984).

[30] Munro, op. cit. (7), 99-101.

[31] Munro, op. cit. (7), 102-7; [David Brewster], 'Parliamentary Report on Lighthouses', Edinburgh Review, 61 (1835), 221-241; note, signed by Alan Stevenson, ibid., 526-8; reply [Brewster?], ibid., 528-31.

[32] Munro, op. cit. (7), 107.

[33] Alan Stevenson, A Rudimentary Treatise on the History, Construction, and Illumination of Lighthouses (London, 1850), 6.

[34] Munro, op. cit. (7), 110.

[35] Munro, op. cit. (7), 109-111.

[36] Stevenson, op. cit. (26), 22.

[37] Stevenson, op. cit. (26), 177.

[38] Roland Paxton, 'Alan Stevenson', Oxford Dictionary of National Biography (Oxford, 2004).

[39] T.1958.49.

[40] T.1958.48.

[41] T.1958.56. All of these instruments now survive in the collections of the National Museums of Scotland, where they were first adequately identified by my colleague Allen Simpson some years ago.

[42] Stevenson, op. cit. (26), 295-6.

[43] Edward Bulwer Lytton, Paul Clifford (London, 1830).

[44] Simpson, op. cit. (1), 18.