The Lalla Rookh, during Thomson’s first winter of ownership, underwent repairs and modifications, and with the guidance of Mrs. Tait was fitted out with draperies, sheets, tablecloths, and dinnerware, sufficient to make it a floating residence for half a dozen people. Thomson playfully applied scientific and mathematical principles to the matter of equipping his vessel. The question of fabric for the bedsheets, he explained to Mrs. Tait, “has, after anxious consideration and consultation with naval experts, been decided in favour of linen. The cotton fabric seems to be too hygroscopic to be suitable for sea-going places.” He wanted the towels and the larger bath sheets to be made of the same material, objecting to the practice that “sometimes bath sheets are made thicker (apparently with the idea of maintaining a constant proportion of thickness to length or breadth) which is a mistake.” Thomson lavished money on his new toy: 12 pairs of sheets, 5 dozen towels, 3½ dozen table napkins of double damask, 10 tablecloths, and whatever quantity of kitchen towels, cloths, and dusters Mrs. Tait deemed necessary. She was to place orders for all these furnishings and advise Thomson of whatever else should be acquired.

Of Mrs. Tait herself, little is known except that she put up with P. G. Tait for many years and bore four sons and a daughter. Tait’s biographer,

C. G. Knott, spends a paragraph describing how Tait had become friendly at Cambridge with two Belfast brothers, William and James Porter, who in their years were third and seventh wranglers respectively. Both tutored for a while at Peterhouse, William eventually becoming master of the college, while James after a time entered the church. A third brother, John, had a distinguished career in the Indian Civil Service. After leaving Cambridge and before returning to Edinburgh, his hometown, Tait was for six years a professor in Belfast, where he maintained close connections with the Porter family. During this time, Knott relates, he “married one of the sisters of his Peterhouse friends.” Which one, what her name was, and on what basis Tait made his selection, Knott fails to say.

By the spring of 1871, at any rate, Thomson was looking forward to sailing adventures as soon as the Glasgow teaching session ended. He urged Helmholtz to come to the British Association meeting at Edinburgh in early August then join him in a scientific party to cruise the Hebrides and western isles on the Lalla Rookh. He proposed to invite not only Maxwell and Tait (if he could be lured away from the golf course) but also Huxley, his opponent on evolution and the age of the earth and sun, and even Tyndall, with whom he and Tait had clashed over the allotment of credit for the laws of thermodynamics. Tyndall had encountered Thomson at Dundee, when the BA met there in 1867. “Thomson met me in the Kinnaird Hall; blocked my passage, smiled and stretched out his hand. I grasped it, expressed in a word my gratification at meeting him, and walked in. Shook hands with Tait afterwards at St. Andrews. They were very cordial to me,” Tyndall recorded in his diary. Thomson easily put any past unpleasantness behind him; probably he could not conceive that anyone might bear a grudge. Even Tait softened a little. A few years later a German astronomer by the name of Zöllner published a Treatise on Comets in which he attacked all and sundry, not only Thomson and Tait but also Helmholtz and Tyndall. Tait approached Tyndall with the most comradely offer he could manage, which was that they should march into battle together. “There will be a splendid row, which is some consolation,” he wrote.

But Tyndall would not be drawn. “Whether it is that the fire of my life has fallen to a cinder, the book has produced singularly little distur-

bance in my feelings,” he replied. “Ten years ago I should have been at the throat of Zöllner, but not now. I would rather see you and Clausius friends than Zöllner and myself. Trust me C. is through and through an honest high-minded man.” Rebuffed, Tait found an opportunity to resume hostilities a few years later. Writing the life of his predecessor at Edinburgh, J. D. Forbes, he disinterred a dispute, dating back to the 1840s, over a theory of glacier motion in which Tyndall and Forbes had been on opposite sides. The subsequent sniping spilled over into the pages of Nature. In September 1873, Tait mocked “the flow of word-painting and righteous indignation which Dr Tyndall so abundantly possesses.” Tyndall shot back, describing Tait as “this man whose blunders and whose injustice have been so often reduced to nakedness, without ever once showing that he possessed the manhood to acknowledge a committed wrong.” The following week the editor referred to further communications he had received from both parties, which he chose not to print as the exchange had “assumed somewhat of a personal tone.”

In Belfast the following year Tyndall served as president of the BA and gave an address that excited controversy and repudiation. Going back to Democritus, Epicurus, and Lucretius, he praised those thinkers throughout history who had striven for rational, material explanations of natural phenomena, rather than resorting to a “mob of gods and demons.” He mentioned Giordano Bruno, burned at the stake by the Church for his heresy in promoting Copernicus and his sun-centered universe. Coming up to date he praised Darwin and the modern biologists who sought to explain the origins of life through science; he even suggested that human beings and their intelligence might have an origin in material processes. He had no specific explanation of how this could come about, but he urged his audience to think of the question rationally, rather than resorting to vague and unquestioning invocations of divine intervention.

This was blasphemy in some quarters. Thomson wrote to his brother-in-law, the Rev. King, that Tyndall’s suggestion was “especially inappropriate.” Stokes said Tyndall was surely wrong: Atoms have no emotion or thought, so how could life made only of atoms acquire such things? Maxwell, typically, produced a poetical satire, beginning:

He then mocked Tyndall’s supposition that collections of atoms could do things that single atoms could not:

(This is a clever play also on Lucretius, who argued in De rerum natura that atoms move at random “for surely [they] did not hold council … flexing their keen minds….”)

To the mechanistic thinkers of the late 19th century it seemed of course impossible, contrary to common sense as well as reason, that inanimate matter could of its own accord turn into sentient creatures. The problem is still not solved: Most scientists today would agree with Tyndall’s proposition that intelligent life can, somehow, arise from inanimate origins, though a number of unimaginative philosophers cling steadfastly to the old view. Tyndall was not, in his specific achievements, a great scientist, yet in his general views he was forward looking and imaginative.¹ Thomson was the opposite. A brilliant scientist and solver of problems, he could not or would not look very far forward, because he did not like to speculate where he had no solid ground beneath him. Therefore he believed that life, especially human beings, came from a divine impulse, and that was that. The combination of technical acuity and imagination in one mind is a rare thing indeed.

***

During these years Thomson adopted an increasingly capacious mode of living. His work on the committee investigating the loss of the

Captain took him to London in the summer of 1871, but now he shunned the train and instead sailed on his new yacht through the Irish Sea and around Cornwall to Dartmouth, Weymouth, and Southampton on the south coast of England, taking the train to London and staying there at the Athenaeum club when necessary but living on his yacht as much as he could. His voyaging had a certain evangelical quality. He persuaded friends and family members to accompany him, cheerily confident they would share his enthusiasm for life afloat. He took his nephew James Bottomley and brother-in-law David King with him as far as Penzance, where King left to be replaced by the other brother-in-law, Bottomley’s father. King said that he would have enjoyed Lalla Rookh more had she remained on the slip at Greenock, and William Bottomley opined (after a tough stint against a stiff east wind) that the best part of yachting was going ashore. Thomson took these remarks as jests and assured Mrs. Tait that everyone had had a splendid time.

He threw himself into sailing with the same energy he used to attack any scientific, technical, commercial problem that came his way. He appeared on deck in the middle of the night to make sure the watches were at their posts. He urged his captain and crew on in conditions they thought dangerous. “You will not rest till you have your boat at the bottom,” Captain Flarty muttered to him once as they fought gale force winds. This he never did, although he managed to run the Lalla Rookh aground some years later, an incident he amusingly recounted to a friend as an experiment showing that wood yielded more easily than rock. Despite, or perhaps because of, his devil-may-care attitude, Thomson was greatly loved by his crew. He talked to the men without pretension, worked as hard as they did, and was eager to learn the arts of navigation. Inevitably, having learned, he wanted to improve and rationalize the science of sailing. When a break from business in London presented itself, he took off for Lisbon and, sailing across the Bay of Biscay, began experiments on sounding devices. Traditionally, sailors threw a weighted line over the side to judge the depth of the water, but for an accurate sounding the ship had to come to a full halt. Thomson set about devising easier, faster, and more accurate methods.

Back in London, between work for the Admiralty and additional duties as examiner for the India Telegraph Service, he arranged a reunion

with several Cambridge friends he had not seen for a quarter of a century. They got together and played a little music, but Thomson found the occasion oddly unsettling. “It was a strange reunion, like a return from another world…. It can never again be what it was, and it is too full of sadness for the present.” The past—Thomson had no time for it. In the few quiet moments he could find, usually when he was alone on the Lalla Rookh, he struggled to compose his presidential address for the British Association’s 1871 Edinburgh meeting, in which he resumed his attack on the geologists and biologists and proposed his cometary idea for the origin of life.

Huxley and Tyndall declined Thomson’s invitation to a postmeeting jaunt on the Lalla Rookh, as did Maxwell, who was preparing to take up his new appointment as Cavendish professor in Cambridge and who was in any case the exact antithesis of the jolly boating fellow. Other duties had prevented Helmholtz from attending the BA meeting, but he came over from Germany later in August to go sailing with Thomson. First, though, he went to St. Andrews where, as Tait had avidly proposed, he might “learn (at its headquarters) the mysteries of GOLF!” Helmholtz failed to succumb. He wrote to his wife: “Mr. Tait knows nothing here besides golf. I had to go along with him. My first swings succeeded, but after that I hit only ground or air. Tait is a curious kind of savage—exists here, so he says, only for his muscles, and only today, Sunday, when he dared not play, though he didn’t go to church either, would he be brought around to rational matters.” In explaining the strange game to his wife, Helmholtz’s scientific precision deserted him: He reckoned that each hole was about one English mile long and that the players walked 10 miles during a round.

Helmholtz went on to stay briefly at Thomson’s new house in Glasgow, while Thomson, out at sea somewhere, arranged where the sailing party should meet. This was a faculty house, part of the new university buildings. Following Margaret’s death, Thomson had not got around to making the place presentable. Helmholtz described to his wife the unfinished rooms, uncarpeted and unpainted, and the furniture stacked here and there, waiting to be set out. It produced in him “an indescribably sad impression, as if no one cared about the place.” In the dining room he came across “an exceedingly fine and expressive portrait of her,

and below it the couch where she used to lie, and her coverlet. I was very sad and could hardly restrain my tears. It is very sad when men lose their wives, and their life is left desolate.” Helmholtz had lost his first wife 12 years earlier, and a couple of years later had married a considerably younger woman. The evidence of Thomson’s loss brought back memories. To him the empty house evoked Thomson’s now empty life—except that Thomson had been so busy equipping his new yacht, sailing to the south coast and to Portugal and back, and attending meetings in London that the state of his Glasgow house had drifted from his mind. Before long he would have to return to Glasgow to start the new session. The house could wait until then.

Helmholtz joined Thomson on the Lalla Rookh at Inverary, on Loch Fyne, where they saw highland games before sailing back to Glasgow and thence to Belfast where they picked up James Thomson. Assorted nephews and sisters-in-law joined and left at various places. They recrossed the Irish Sea and sailed about the west coast of Scotland until they reached the Blackburns’ house at Rushven, on the Moidart peninsula. Helmholtz found this “a lonely property, a very lovely spot on a bay between the loneliest mountains.” Jemima Blackburn’s animal and bird drawings impressed him. She painted a little watercolor of him and the Thomson brothers observing birds out at sea. The Thomsons, a couple of nephews, and the boisterous Blackburns and their children constituted a “friendly and unconstrained” party. Helmholtz was taken aback by Thomson’s habit of abruptly withdrawing from the games and conversations, to sit with his green notebook and make calculations. “How would it be if I accustomed the Berliners to the same proceedings?” he asked his wife, with puzzlement or perhaps envy. Oddest of all, Helmholtz thought, was that after Thomson had assembled his guests at dinner aboard his yacht the night before they were to set off toward Skye, he immediately disappeared to his cabin to work at some problem in his notebook. The dinner conversation faltered, and Helmholtz went off to stroll up and down the gently rolling deck with as much “unsteady elegance” as he could muster. He observed, of himself or Thomson or both, that “a husband who is no longer in his first youth feels uncomfortable when he wanders about in the world, all by himself, without higher guidance, and I think that if the

world were peopled with men only, it would not be particularly beautiful, but would be very practical, and not at all refreshing.”

For Thomson nothing could be more refreshing than immersing himself in practical questions. He loved the sea as much as he loved natural philosophy, perhaps more. The six winter months in Glasgow he taught and lectured and worked on scientific matters. The summer months he arranged to spend as much time as possible on the Lalla Rookh, sailing as often as he could but relying on it too as a refuge from the endless demands on his time. Problems of navigation gave his intellect a full range of scientific and practical matters to attack. Merely being at sea presented interesting questions. He had long corresponded with Stokes about the numerous physical and mathematical issues arising in the motion of fluids: streamlines, waves, turbulence, eddies, rotation, and so on. Underneath all these phenomena lay Newton’s simple laws of motion, time-tested and elementary, but in fluids these laws manifested themselves in an enormous variety of ways. Fluids, treated by mathematical physics, could be idealized as incompressible or more realistically given some degree of compressibility. They had density, viscosity, elasticity, all of which could change with temperature. Thomson loved this kind of thing. He had no fear of mathematics and no particular aesthetic sense of it either. If he found that a model of some problem wasn’t yielding the full range of observed behavior, he would happily pile on more complications. There is an old joke about a theoretical physicist asked to come up with improvements in milk production on a dairy farm, who after months of secretive analysis announces that he has solved the problem in its entirety, at least for the ideal case of a perfectly spherical cow. Thomson wouldn’t have understood the joke. What do you mean, a spherical cow?

While sailing, he and Helmholtz had indulged in a sort of competition to see who could correctly explain the behavior of various waves and ripples they observed from the deck of the Lalla Rookh. When Thomson had to leave the yacht for a while to attend to some problem ashore, he said jocularly, or perhaps not, “Now, mind, Helmholtz, you’re not to work at waves while I’m away.” A later commentator disparaged Thomson as someone who “had immense intellectual strength, but was deficient in intellectual taste.” How could a man capable of founding thermodynamics and laying the groundwork for modern electromagnetic theory fritter

away his time and mental energy in explaining the ripples on the Sound of Mull, or tinkering with devices to measure the depth of the ocean from a moving boat? But to Thomson a problem was a problem was a problem. Whatever puzzle came before him engaged his interest. In a sense his contributions to thermodynamics and electromagnetism were the aberration, his interest in telegraphy and navigation the more characteristic examples of his talents. By resolving certain contradictions between Faraday and the French electrical theorists, and between Carnot and Joule’s views of heat and work, he had found a way forward, illuminated the path that theoretical physics must take. But having gotten over the immediate difficulty, Thomson’s attention turned elsewhere. It was others, in the end, who completed the journey. Thomson was a practical thinker, a resolver of difficulties, not a metaphysician.

***

On his first voyage with the Lalla Rookh through the Bay of Biscay, Thomson had experimented with a sounding device consisting of a 30-pound lead weight attached to a reel of thin piano wire that could be spun with ease from the stern of the ship. The design owed something to his experience with the machines that played out telegraph cables. An accurate sounding demanded that the ship come to a halt, as before, but Thomson’s idea was that the depth could be taken much more quickly than with a weight on a hemp rope, so the interruption to progress would be minimal. His first attempt almost came to grief because of an elementary difficulty that he was “much ashamed” not to have thought of beforehand. The wire unspooled nicely, and he got a rapid and accurate depth—2,600 fathoms, agreeing with the chart. But when about a third of the wire had been retrieved, the reel showed alarming signs of strain and began to buckle. Even though the tension on the wire was at most 50 pounds, Thomson realized, the effect was additive: Each turn of the reel added that much tension, so that if the whole length were wound in, over 100 tons of pressure would squeeze the reel.² The crew had to stop and haul in something like a mile of thin wire by hand. Thomson devised a

secondary pulley that would take up the tension and allow the wire to be retrieved, another trick that came from cabling expeditions.

Two years later, sailing to Brazil on the cable ship Hooper with his old telegraph colleague Fleeming Jenkin, Thomson found that he could take “flying soundings” with reasonable accuracy. With a light wire unreeling freely, he assumed that the weight dropped vertically from the point of release while the ship steamed on. The ship’s speed being known, application of Pythagoras’s theorem provided the depth from the distance traveled and the length of wire played out. A table converting length of wire and ship’s speed into depth relieved the sailor from needing to know his square roots.

Thomson took out a patent on his sounding machine in 1876, but soon afterward started work on a different system. Instead of a passive weight, he attached a simple pressure gauge to the end of the wire. Essentially this was nothing more than an open-ended glass tube with some air in it, weighted so as to keep the open end facing down. As it descended and pressure increased, water advanced up the tube, squeezing the air into a smaller volume. To mark the water level, Thomson tried dyes that got washed away as the water rose, but eventually he settled on a reactive chemical that changed color on contact with water. When the pressure gauge was retrieved, visual inspection revealed how far the water had made it up the tube, which indicated the greatest pressure it had experienced. This yielded directly the maximum depth attained.

This pressure-recording device was nothing new, and the patent for the chemical marker belonged to a T. F. Walker. But Thomson, with his piano wires and pulleys (adapted from cabling machinery), supplied the means to raise and lower the device easily and reliably. Again, Thomson showed his knack for putting together disparate elements, solving some practical difficulties, and coming up with a working system—an empirical counterpart to his theoretical achievements in thermodynamics and electromagnetism. A series of patents in 1880, 1883, and 1885 stamped Thomson’s name on the system, and although bureaucracy proved sluggish and reluctant, he succeeded in getting his device adopted by the Royal Navy and other navies. Only with the advent of echo-sounding sonar devices in the early 20th century did new technology supplant his basic design. Of course, there were licensing fees, royalties, and consult-

ing opportunities. Thomson spent many long days aboard the Lalla Rookh making sure his system was both reliable and practical for the average sailor. He took pride in being meticulous as well as ingenious, and when the thing was ready he was equally assiduous in maintaining his legal rights and establishing his income.

The 1873 trip to Brazil had another satisfactory outcome for Thomson. To repair electrical flaws in cable coiled aboard the Hooper, the expedition paused for a couple of weeks on Madeira, one of the Canary Islands off the west coast of Africa. There Jenkin and Thomson made the acquaintance of Charles Blandy, a local businessman and prominent islander. (To this day the Blandy company produces madeira and other fortified wines.) Blandy had two daughters, who learned Morse code from the visiting technical men. The women signaled with a lamp from their house to the Hooper, anchored a mile and a half distant.

Jenkin always remembered Madeira with a pang of alarm. While they were out one day riding on the steep island hills, Jenkin’s horse darted unexpectedly and almost pushed Thomson’s over the cliff edge. “No harm was done,” R. L. Stevenson relates, “but for the moment Fleeming saw his friend hurled into the sea, and almost by his own act; it was a memory that haunted him.” Jenkin idolized Thomson almost to the point of worship and could hardly bear to think he might have killed him. Thomson, however, never mentioned the incident and apparently thought no more of it. The moment passed, and he was safe. It was past; no reason to dwell on it.³

Thomson, by contrast, left Madeira with a budding romance. Anecdote has it that when the Hooper steamed from the island to lay a cable to Brazil, Thomson’s attention was drawn to a white cloth fluttering from a window of a house overlooking the port. Peering through his eyeglass, Thomson interpreted the Morse code flapping: “Goodbye, goodbye, Sir William Thomson.”

The message drew Thomson back the following May, as soon as the Glasgow session had ended, in the Lalla Rookh. He and Frances Anna Blandy, known as Fanny, were married on June 24, 1874, in the British Consular Chapel on Madeira. Thomson was two days shy of his fiftieth birthday. His new wife was in her mid-30s. Though little is known about either his first wife or his second, they were clearly very different. Margaret Thomson had been, even before her years of ailing, a refined, sensitive, artistic soul. Fanny Thomson was, like her husband, cheery and gregarious, and unlike him socially accomplished and elegant. She was a capable, outgoing, practical woman. She loved to organize dinner parties, attending to flowers, seating arrangements, menus, and all manner of diversions. She would gently kick her husband’s shin under the table if he seemed about to reach into his pocket for a green notebook.

Thomson’s chance encounter the previous year evidently awakened feelings that his frenetic activity had concealed even from him. The day after Fanny accepted his proposal he wrote to his sister: “When I came to Madeira in the Hooper it had never seemed possible that such an idea could enter my mind, or that life could bring any happiness. When I came away in July I did not think happiness possible for me, and indeed I had not begun even to wish for it. But I carried away an image and an impression from which the idea came…. Hope grew stronger till yesterday, when I found that I had not hoped in vain…. When you know Fanny you will be able to really congratulate me. Even now I think you will be glad for my sake.” No one ever remarked of Thomson, at any time in his life, that he seemed to be an unhappy or melancholic or brooding kind of man, but his long years with Margaret had been mostly toil and worry. Fanny was a bright soul and charmed Thomson’s brother and sister and all the nieces and nephews. She traveled with him frequently and made her own social arrangements while he attended scientific or business meetings. Thomson’s increasing wealth and reputation made him the center of a widening circle of notable acquaintances in business and politics as well as science, but Thomson himself had little time for purely social matters. Fanny gave him a life appropriate to his circumstances, and he gratefully participated.

Later that year the newlyweds bought a piece of land at Netherhall, on the Ayrshire coast near Largs, where Thomson had so often spent

summer months. He undertook the design and construction of a splendid house.⁴ He supervised carpenters and masons, with the result that everything took longer than it need have done, as he could never resist the urge to improvise. After the house was finished he used it as a venue for experiments in domestic science. It was probably the second house in Great Britain to be equipped with electricity (the first was Cragside, near Newcastle, built by the industrialist Sir William Armstrong, who had a dynamo installed at a waterfall on his property). Thomson at first used large storage batteries of a French design, similar to a modern lead-acid car battery, which he wrote enthusiastically about to the Times. By the early 1880s he was experimenting with generators running from the domestic gas supply and running an impressive variety of incandescent lamps as well as electrical experiments.

With his new wife, new yacht, and new house along with all his teaching, research, and commercial activity, Thomson displayed a seemingly infinite capacity for doing things. A green notebook accompanied him at all times. Even before his marriage he had begun to rely on a string of assistants, former Glasgow pupils, who helped him in the lab and with his writing. He explained to one colleague, “as I have so many engagements, and so much laboratory work, that I am kept constantly standing and walking about, I can seldom sit down to write anything, and am obliged to do nearly everything I wish in black and white by dictation.” Early in 1874 he gave his presidential address to the Society of Telegraph Engineers, then in its third year. He spent a couple of weeks working with his assistant in odd moments and came up with four minutes’ worth of material. When it came time to deliver his lecture, he started with the few sentences he had painfully composed, then winged it. A stenographer took down his talk and the lecture reads loosely but is cogent and lively. C. G. Knott on one occasion was with Thomson and Tait in Edinburgh and had agreed to write up Thomson’s remarks to the Royal Society of Edinburgh for publication in Nature. He had difficulty sum-

marizing Thomson’s largely ad lib presentation and approached him the next day for further enlightenment. Thomson stared perplexedly into space for a while, struggling to recollect, then had a sudden thought: “Oh, I’ll tell you what you should do. Just wait till the Nature report is published—that fellow always reports me well.”

For all the evidence of the dissipation of his intellect into innumerable half-completed researches, Thomson showed an unswervable ability to pursue technical and practical projects to a fine state of perfection. In a letter to the Times and in contributions to the British Association, he had suggested that each lighthouse should signal with a distinctive Morse code pattern, as he had found that sailors coming across a light were frequently so unsure of their location that they didn’t know which hazard they were near. He badgered naval men and civil servants whenever he had an opportunity, and his system was eventually adopted.

Since the late 1860s, Thomson had busied himself with an analysis of tide heights at various ports. This was important information for the Admiralty as the Royal Navy stationed itself across the globe, and the British Association had taken on tide prediction as an official project, with Thomson taking a lead role. It had been established that tides at any location could be analyzed into a series of harmonic components, each component having a certain period and a magnitude, from which tides could be predicted with good accuracy. This entailed complex mathematical analysis of measured tides and further mathematics to make a prediction. Around 1876, Thomson devised his Tidal Harmonic Analyser, essentially a mechanical calculator, based on an invention by his brother, and using a set of cogs of appropriate sizes to mimic the components. With the machine correctly set, for a specific location, anyone could predict future tides by cranking the handle.

The Analyser was another characteristic invention. The mathematics it embodied came from others, notably Laplace in France and Airy in England. The germ of the mechanism came from James Thomson. But it was William Thomson who combined the theoretical and practical elements, recast the mathematics into amenable form, developed his brother’s innovation into a more general calculating device, and produced a working machine that did exactly what it was supposed to do, in a way that demanded no expertise on the part of the operator. It led him also

into a petty dispute in which Thomson showed his increasingly inflexible assertion of his own priorities and interests.

Since about 1872, Edward Roberts of the Nautical Almanac Office had assisted Thomson on the tidal prediction project by performing the tedious but routine calculations needed to obtain the magnitudes of harmonic components from observations at various ports around the world. When the mechanical calculator came to be built, Roberts had the responsibility of working out such details as the correct numbers of teeth for the gears. In 1879 he composed a short paper for the Proceedings of the Royal Society, with the title “Preliminary Note on a New Tide-Predictor.” This came to Stokes, in his editorial capacity, who learned indirectly that Thomson wished the title to be changed to “Preliminary Description of Sir William Thomson’s Tide Predictor Constructed for the Indian Government.” Stokes then related to Thomson how “utterly surprised” he was that this “very mild and unobjectionable” change caused Roberts to fly into a huff and refuse to have the paper published in its new form. In the end it was published with its original title, but then Roberts began to speak of the Roberts tide predictor and claim that the important part of the invention was his, acknowledging Thomson only for one or two useful hints. As Thomson explained the matter to Stokes, on the other hand, the design was due to him, except that while riding on a train from Brighton to London, a Mr. Tower with whom he was traveling had suggested driving the machine with a chain-and-pulley mechanism originally devised by Charles Wheatstone for his old letter-printing telegraph receiver. “That is the very thing for me,” Thomson had instantly said.

The only innovations Roberts introduced into the predictor were bad ones, Thomson said, which he had to take out again. He wished, in the end, he had had the machine made by James White in Glasgow, a superior instrument maker and a man he could trust. He engaged White anyway to then start work on a predictor with additional improvements, so there would be no doubt who was the true inventor. This little flap amounted to nothing much, but any of the hesitation and deference Thomson had shown 20 years earlier in dealing with Whitehouse’s claims over telegraph theory and instrumentation had long since vanished. Thomson picked up clues and hints wherever he could find them and

relied on assistants and engineers and technical men to help him refine his ideas and turn them to practical use. But when a finished product was ready for display to the world, there was no question whose name would be attached to it.

***

The sounding machine, the tide predictor, the signaling code for lighthouses—estimable innovations all, but inferior both in importance and in the magnitude of bureaucratic struggle they entailed to the central element of Thomson’s career as a marine philosopher. The essential navigational device, passed down from antiquity, perhaps originally from China, was the compass. At the beginning of the 19th century boats were almost wholly wooden. First, metal rivets made an appearance, then a few strengthening iron beams were incorporated into hulls. Brunel’s steamer, the Great Britain, launched by Prince Albert in 1843, was the world’s first fully iron-clad ocean-going ship. Commercial shipping rapidly changed from wood to steel. The Royal Navy followed suit, as slowly as it could decently manage. Iron ships still relied on compasses, but iron had magnetic properties of its own. Ships ran aground because their compasses no longer pointed north but were deflected by the iron hulls and superstructure that carried them. The problem was not new in Thomson’s day, but the solutions devised thus far he found unsatisfactory. Naturally, he could do better.

Maritime legend has it that a Portuguese captain, João de Castro, noticed in 1538 that his compass needles twitched when heavy iron cannons were moved about the deck. Two and half centuries later William Wales, an astronomer sailing with Captain Cook’s Resolution in the south Pacific in the 1770s, noticed the irresolute behavior of the ship’s compasses but failed to see the cause. Wales reported that the compass direction drifted depending on the ship’s course, its latitude, and the placement of the compass on the ship, but concluded somewhat obtusely that there must be something wrong with the compasses. Finally, on Royal Navy expeditions to Australia in 1798 and 1801, Matthew Flinders systematically studied compass deviations and began to understand their origin. He noticed particularly that the departure of a compass needle from magnetic north changed sign when the ship crossed the equator—that is, it

erred to one side in the northern hemisphere and to the other in the south.

Flinders had more time to ponder this problem than he might have liked. Unaware, sailing the south seas, that France and England were at war, he was captured by the French in 1803 and remained a prisoner of war for several years. With ample opportunity for reflection, he came to the conclusion that compass deviation was related to the “dip” of the terrestrial magnetic field. If the earth is pictured as a giant bar magnet, its magnetic poles coinciding approximately with the geographical poles, then lines of magnetic force will emerge vertically at the poles and curve around the planet, becoming parallel to the earth’s surface at the equator. The angle between the magnetic field and horizontal is the dip.

War being somewhat more gentlemanly in those days, at least for officers, Flinders was able to publish his findings in the Proceedings of the Royal Society while still imprisoned. His analysis attracted the interest of scientists but not naval men. Freed in 1810, he proposed correcting a compass by placing an iron bar adjacent to it, finding the correct position by trial and error. His thinking was that the iron in any ship, though it was scattered about in some complex pattern, would act to a first approximation like an iron rod at some fixed location relative to the compass. A properly placed “Flinders bar,” as it eventually became known, would cancel the magnetic distortion produced by the ship, leaving the compass to measure the true magnetic field of the earth.

Flinders died in 1814, at 40, having been unable to submit his correction bar to practical scrutiny. Six years later Peter Barlow of the Royal Military Academy tried a similar scheme involving an iron plate positioned near the compass. Barlow’s system went into practice on a number of ships but didn’t succeed widely, in particular because a plate that corrected compass error north of the equator was found, perplexingly, to magnify it in the southern hemisphere. In the meantime other scientifically inclined navigators had confirmed and extended Flinders’s original analysis of the variation of errors with a ship’s course and latitude, without so far coming up with any systematic account of the cause or a practical method to deal with the deviations.

There arose the practice of “swinging” a ship to quantify compass deviation. In a suitable harbor, a captain would use geographical land-

marks to align his ship north, south, east, and west, and at a dozen or more compass points in between, so as to measure the discrepancy between the known direction and the compass indication. This procedure yielded a table of corrections which the navigator then applied, out in the open sea, to get true direction from compass reading. This assumed that the necessary correction, measured in one place, would work anywhere on the globe, which Flinders and others had already shown not to be the case. The Royal Navy made it official policy to use correction tables rather than the Barlow plate and other unreliable devices, with the recommendation to captains that they should swing their ships regularly, especially in the course of long voyages. This itself was no easy matter: Swinging a ship was time consuming and required some independent way of establishing compass directions. Nor was the use of correction tables as straightforward as it might appear. Mistakes happened when a navigator subtracted a tabulated correction from a compass reading, instead of adding it—not as absurd as it sounds, as the problem is much like puzzling out whether to put clocks forward or back when going from summertime to wintertime, without the benefit of a handy mnemonic, and with the rule being different from one ship to another.

By the mid-1830s, compass deviations were poorly understood but were undeniably getting larger as ships used increasing amounts of iron in their construction. No satisfactory mechanical correction existed, and correction tables had a way of confusing all but the most sophisticated seamen. Ships ran aground with staggering frequency—hundreds of British vessels, naval and commercial, were lost every year—and the common sailor learned a great distrust of compasses of any kind (which, making matters worse, were often poorly built and unreliable from the outset, apart from the question of deviations).

In 1835, on his own initiative, Captain Edward Johnson of the Royal Navy investigated compass deviation on the Garryowen, a 130-foot-long iron paddle steamer with an enormous funnel 28 feet high. He placed compasses at many points around the boat and swung it to measure deviation at different locations. Imaginatively, he also set a number of compasses around the harbor where he was swinging the Garryowen and found that these too suffered deviations changing with the orientation of the ship. This was a crucial though dismaying discovery. Flinders, Barlow,

and others had taken it for granted that an iron ship passively distorted the magnetic field passing through it. Johnson now concluded that in addition an iron ship had some permanent magnetism of its own, which is why it affected compasses nearby. He speculated that as iron was heated and cooled and shaped and hammered, immersed all the while in the earth’s magnetic field, it acquired permanent magnetism that was then built into the ship under construction.

The British Admiralty finally lumbered into action, forming in 1837 a Compass Committee to address this “evil so pregnant with mischief”—namely, the dismal performance of compasses on Her Majesty’s warships. (Queen Victoria came to the throne that same year, which perhaps provided a little fillip of opportunity for change and reform in the realm of officialdom.) The Committee investigated all aspects of compass performance, including basic design and quality control as well as deviation. Captain Johnson served on the Committee, which the following year asked Astronomer Royal George Airy to investigate deviation on the paddle-steamer Rainbow. Airy, a man of great mathematical skill but not altogether conversant with magnetic phenomena, concluded that compass deviation on the Rainbow came almost wholly from the ship’s permanent magnetism, and worked out how to compensate for this interference by positioning two small bar magnets, one on the fore-aft line through the compass, the other laterally—athwartships, in nautical language. Airy’s method was simple. The ship was swung to point north, and one magnet was placed to make the compass point north as well. The other magnet was positioned similarly by swinging the ship east-west. (In fact, Airy proved that a single magnet would correct the compass, but working out its location required a difficult mathematical analysis.)

The technique seemed to work reasonably well, and commercial shippers (who showed more enthusiasm than the Admiralty for solving the problem, being acutely aware that days lost in passage from navigation errors translated into lost business) hired Airy to install his correction system on a number of ships. But the vessels he corrected were mainly of wooden construction, although with a growing number of iron components. As the amount of iron in ships grew larger, compass deviations grew too, and Airy’s method proved insufficient. In particular, when a ship corrected in England went south of the equator, its compass fre-

quently became less trustworthy than if it had not been corrected at all. The reason is that permanent magnetism is not the whole story. A mass of iron distorts a magnetic field that passes through it, in a way that depends on geometry and orientation. The simplest way to deal with this phenomenon is by thinking of the iron developing an induced magnetism in response to the external field. An iron ship, therefore, acts on a compass in two ways: There is a fixed, permanent, or hard magnetism and a variable, induced, or soft magnetism that depends (as Flinders had long ago found) on the ship’s position relative to the earth’s magnetic field—in other words, on its latitude and heading.

As if this were not complication enough, there is also heeling error. When the ship tilts to one side or the other, out of vertical, its geometry relative to the earth’s magnetic field changes, and so the induced or soft magnetism changes too. As Airy had calculated, the hard magnetic error in a compass can be fixed with a couple of permanent magnets, suitably placed.⁵ Likewise, the soft error can in principle be compensated by placing soft iron correctors, whose induced magnetism counteracts that of the ship as a whole. But the heeling error, which became significant only for fully iron-built ships, introduces additional complication in placing the correctors. Heeling error, moreover, is linked to the variation of error with latitude, since both depend on the angle with which the earth’s magnetic field passes through the ship.

Although Airy was able to earn handsome fees for correcting compasses (£100 or more per ship, compared to his annual salary as astronomer royal of £500), he showed no enthusiasm for going into this line of business as the magnitude of the task became more apparent. He acknowledged the importance of soft as well as hard magnetism and experimented with adding suitable iron correctors as well as magnets, but then largely lost interest. Such complications, on the other hand, were posi-

tively an attraction to William Thomson. It is not clear when he first took a serious interest in compass correction. A letter to Stokes as early as 1850 tells of him going to Borley Rectory in Suffolk to visit the Rev. W. W. Herringham, an old Cambridge friend, where he planned “to see ‘the Retribution’ swing, for det^g the devⁿ of his compass.” On the 1858 Atlantic voyage, the mass of cable aboard the Niagara had upset the compasses enough that it steamed away from the midocean rendezvous on the wrong bearing and had to change course and follow one of the smaller ships instead. Thomson, however, recorded no recollection of this incident.

In 1871 the editor of Good Words once again called on his friend for an article on some technical or scientific subject, and Thomson, proud new owner of the Lalla Rookh, decided to write about the nautical compass. His account, a general history of the compass, appeared in 1874 and did not offer any great novelty. But that same year brought the death of his old friend Archibald Smith, Glaswegian, London lawyer, and sometime mathematician. Thomson wrote an obituary notice of him for the Royal Society. Smith was just over 60 years old when he died, and Thomson blamed his demise in part on exhaustion. While working at the law during the day, he devoted his evenings to mathematics, and the particular problem that consumed so much of his energy was a complex and detailed analysis of compass deviation. This was official labor, undertaken for the Admiralty’s Compass Committee. When Airy’s compensation system proved inadequate, the Navy settled on a policy of mathematical correction but in a more sophisticated way than before. The soft iron deviation varied, as was now clear, with the ship’s bearing, because of the dependence on angle between ship axis and magnetic field. The old system of swinging a ship to get a single set of corrections was utterly inadequate, but Smith (elaborating an earlier treatment by Poisson) devised a method for determining a ship’s soft iron properties from a prescribed set of swings, which could then be transformed into a mathematical formula to derive tables of compass corrections, individualized to each ship, and varying according to the ship’s bearing.

This was a fantastically intricate business. Every year ships’ captains had to perform a complex set of swings to obtain the necessary data. These numbers went to London, where Smith’s mathematics transformed them into an array of correction tables that were sent to the ship con-

cerned for a navigator to apply to the ship’s compass. The mathematics, to Thomson’s eyes, was refined, elegant, and powerful, but as a system to allow an ordinary sailor to chart a course, the method was beyond impossible. The Admiralty rather preferred it this way. It wouldn’t do for plain sailors to know how to steer their ships; that was a task for officers. Only a select few could master the art of compass correction, which obviated the danger of untutored seamen going astray or worse through blind trust in a poorly corrected compass. On the other hand, the mathematical correction method offered so many chances for error that it was far from foolproof, even in expert hands.

Between sailing his own ship, writing for Good Words, and studying Smith’s handiwork, Thomson found a new challenge to latch on to. His demotic instinct rebelled against the mathematical system in favor of the kind of mechanical compensation that Airy had tried, which in principle put navigation in the hands of everyday sailors. In 1879 he wrote another article to summarize his progress but admitted the task was enormous. “When I tried to write on the mariner’s compass,” he told his readers, “I found I did not know nearly enough about it. So I had to learn my subject. I have been learning it these five years, and still feel insufficiently prepared to enlighten the readers of Good Words upon it when I now resume the attempt to complete my old article.”

Apart from the difficulties presented by the hard and soft magnetism of iron ships, Thomson discovered, there was a woeful history of compass design that ignored elementary matters in dynamics. A compass needle is a magnetized rod, which tries to align itself with the earth’s magnetic field. Centuries ago, compass designers had learned to mount the needle on a support or card that floated in a bowl of water—not a practical solution for a ship rocking around in a violent sea. Instead, a compass card balanced on a pivot—a so-called dry card—was set in a bowl and mounted in gimbals (a 16th-century innovation) that allowed it to rotate freely on perpendicular horizontal axes.

A single needle sitting on the diameter of a card does not work well. In a vessel rolling side to side, this needle experiences a purely dynamical influence that tries to line it up with the axis of the ship, so that the needle twists about its own long axis. Mariners often made their compass needles heavier, thinking they would be more stable. But the heavier the

needle, the more it responds to dynamical rather than magnetic forces. A long, heavy needle on a fiercely rolling ship will reliably and stably point toward the bow, regardless of where the ship is heading.

The Compass Committee settled on a design in 1840 that dealt fairly well with these problems. On the compass card four needles were mounted parallel, some distance apart, and placed in such a way as to the make the card symmetric in its dynamic properties (that is, it would rotate with equal stability about any diameter). “By a happy coincidence,” as Thomson put it, this arrangement of needles also had beneficial magnetic properties. A compass in an iron ship experiences the earth’s magnetic field, distorted by the body of the ship. To a first approximation, the distortion pulls a needle out of true, to one side or the other. But there are also higher-order disturbances, with more complex geometry—harmonic components, essentially, of the distortion. With multiple needles correctly placed, the most important of these higher order effects disappear, because they exert equal but opposite pulls on the different needles, leaving the compass card as a whole unaffected.

This card, of mica-covered paper 7.5 inches in diameter, with two needles 7.3 inches long and two 5.4 inches long, formed the heart of the Admiralty Standard Compass. Considerable thought had been expended on the design of the pivots, the compass bowl in which the card was suspended, the gimbals, and so on. It was, by deliberate choice, an uncompensated compass. It came with instructions to place it on a ship’s midline, as far away as possible from any large iron structures. Ships’ masters received detailed education on how to swing their vessels and how to use Smith’s correction tables. In 1842 the Admiralty set up a Compass Department, under Johnson, to oversee everything from manufacture to testing to installation to maintenance.

The Admiralty Standard Compass was therefore not merely an instrument but an entire system, with detailed rules and regulations and, inevitably, an attendant bureaucracy. In its day it represented an enormous leap forward in quality and reliability, and many other navies around the world adopted it. During the mid-1840s hundreds of the new compasses were ordered and installed.

The lurch into modernity signified by the Admiralty Standard Compass did not pull the rest of the Royal Navy along in its wake. In particu-

lar there was resistance to the commissioning of fully iron-built steam ships. At first, legitimate questions arose about the reliability of steam technology and the soundness of riveted metal hulls, but these subsided as commercial shipping interests forced the rapid development of better engines and more robust ships. During the 1840s steam and iron drove out wood and sail in the merchant marine, but now the old guard of the Admiralty resisted for sentimental reasons. Some sections of the press chimed in too, declaring that the splendidly rigged wooden ships that had built and now safeguarded the British Empire should not be thrust aside to make way for ugly, smelly metal boats.

Consequently, the Admiralty Standard Compass became part of entrenched naval practice when iron ships had not yet put in an appearance, so that compass errors were generally small and mathematical correction worked reasonably well. By the early 1850s, when the Admiralty could no longer resist steam and iron, the Standard Compass system was inviolate, although difficulties in dealing with much greater compass errors were by then becoming apparent. Deviations of one or two degrees, as in the older ships, posed no insuperable difficulty. Deviations of 10 or 20 or more degrees, which were beginning to be found, were another story altogether.

Captain Johnson had long ago seen these difficulties coming and urged a reconsideration of Airy’s correction methods. But he died in 1853. His successor, Captain Frederick Evans, had no scientific training but by dint of great effort mastered the mathematical correction of Archibald Smit—and having mastered it, resisted any hint of change. In any case Airy’s method had failed to prove its worth. In 1854 Airy had compensated the compasses of the Tayleur, an iron-clad passenger vessel. On its maiden voyage, one day out of Liverpool, it ran aground in heavy seas with the loss of 290 out of 538 people. At the British Association meeting in Liverpool later that year, with the tragedy still fresh in the public mind, the Rev. Dr. William Scoresby, formerly captain of an arctic whaler, later churchman and amateur scientist, charged that Airy’s compensation methods were not to be trusted because a ship’s magnetism could change. Just as banging and hammering during construction imparted permanent magnetism to a ship’s iron components, Scoresby said, so violent motion in a storm could do the same. This was a highly dubious conten-

tion, even allowing for the rudimentary understanding of magnetism at the time, and Airy argued against it. But to nonscientific observers, including Captain Evans and many Admiralty figures, Scoresby’s alarms were disturbing. Evans oversaw a test of Airy’s compensation methods on a smaller vessel, but despite a good result remained unconvinced and stuck to Archibald Smith and his mathematical tables.

The loss of the Tayleur and the ensuing controversy at the 1854 BA meeting caused a ruckus among insurers, who became reluctant to offer coverage for new ships. The BA in response formed the Liverpool Compass Committee, which reported in 1860 with a broad recommendation in favor of compensation rather than mathematical correction, though it was clear both systems had drawbacks. Evans conceded that some basic correction by magnets or soft iron had merit, though he hesitated at “the placing of so dangerous a tool as a moveable magnet in the hands of the untrained navigator.” He and Smith produced a series of immense Admiralty manuals on the mathematical correction of compasses, which found favor with many navies around the world. Even a simplified version, though, was too forbidding for the average mariner.

In the meantime iron ships were getting bigger, and in what Thomson, with a jab at Darwinism, called “a process of ‘Artificial Selection’ unguided by intelligence” compasses got bigger too. Highly decorated compasses with long needles looked mighty impressive on the gleaming bridges of new ocean liners, but the bigger the needle, the greater the magnetic force needed to keep it aligned. Dynamical problems resurfaced. The big compasses tended to be unstable or else unresponsive. Worse, a bigger needle needs a stronger magnet to correct it, to the point where secondary magnetic interactions between needles and compensators added to the local magnetic distortions. As one naval history put it, “between 1850 and 1880 ships were, therefore, sailing about with unsteady compasses and heavy deviation tables, and the officers were blaming the compasses instead of mastering the real enemy—inadequate compensation.”

William Thomson, coming across all this in the 1870s, found a problem to relish. A handful of elementary and unarguable physical principles dictated the dynamics and magnetic behavior of compasses. The interaction among the earth’s magnetic field, the hard and soft magnetism of a

ship, and magnetized needles entailed huge mathematical complexity but no scientific novelty. At the end of it, there was a practical issue to be dealt with, and the solution should be one that ordinary seamen, not just mathematically trained navigators, could grasp and use. In 1876, after several summers of experimenting on the Lalla Rookh, Thomson took out the first of several patents for a compass design of his own. Like his sounding machine and tide predictor, Thomson’s compass contained little that was truly original. It was, like the Admiralty Standard, intended as a complete system, but unlike the official instrument it was intended for the use of ordinary sailors. Thomson, one of very few knowledgeable sailors who could actually appreciate the exquisite mathematics of Smith’s correction methods, shunned the mathematical system in his own compass design.

As early as 1874 Thomson had written to Evans suggesting that a lighter compass card, properly suspended, would be more stable in heavy weather, and with guns firing, than the Admiralty Standard Compass card. Evans’s cool reply, on top of the difficulty he had experienced in getting his sounding machine tested, caused Thomson to remark later that “innovation is very distasteful to sailors. I have a semi-official letter to that effect.” This wariness set the tone for subsequent battles.

In his 1876 compass, Thomson turned the compass card into a light aluminum ring, 10 inches across. Putting all the weight at the edge gave the ring an extraordinarily long period of natural oscillation, almost 60 seconds in the prototype, whereas the Admiralty Standard card had a period of perhaps 20 seconds. As ships got larger, they rolled more slowly, and Thomson perceived the importance of giving the compass card a slower oscillation than the ship itself—otherwise in heavy seas the compass would tend to rock in synchrony with the ship and become magnetically unresponsive.

Within this ring Thomson then suspended eight short, slender needles on silk threads. His card was lighter and magnetically more sensitive than the Admiralty Standard, but at the same time had greater dynamical stability. Its magnetic delicacy made it easier to compensate. As well as using magnetic and soft iron correctors more or less as Airy suggested, Thomson included a Flinders bar (something the Liverpool Committee had strongly endorsed) because it largely took care of heeling and

related errors. Characteristically Thomson designed a complete compass mounting or binnacle⁶ in which all the correctors could be positioned in a prescribed and restricted fashion, so that any sailor with a few hours of training could compensate the compass in a reliable way. Some naval historians have charged that Thomson gets too much credit for merely putting Airy’s method into practice, but as well as adding the Flinders bar Thomson took pains to devise a system in which compensation would be both straightforward and trustworthy. Captain Evans of the Compass Department had complained with some justice that Airy’s prescription was too loose to yield consistent results; there was insufficient guidance on the size and strength of magnets and correctors, how close they should be to the compass, and so on. In Thomson’s compass the correctors were designed along with the card and so had optimal properties and placement to harmonize with both the magnetic and the dynamical attributes of the compass card. Thomson “enunciated no new principles but was the first to combine successfully all the requirements in one compass and binnacle,” according to one history of navigation, and this was no small achievement.

Airy was unimpressed. One of Thomson’s students took an early, admittedly rather crudely mounted compass to him at the Royal Observatory but Airy, after looking at it for a while, just said, “It won’t do.” The mandarins of the Admiralty were likewise indifferent. The superintendent of compasses by this time was William Mayes, who had sailed with the Agamemnon on Atlantic cabling expeditions. Evans had by now ascended higher but maintained overall control of compass matters. Complaints were coming in, with increasing frequency, from commanders who found that the Admiralty Standard Compass performed poorly in ships moving at speed or when firing guns. But Mayes, with Evans’s backing, resisted all thoughts of change or innovation. Their job, they said, was to implement policy set out in Admiralty documents.

By the late 1870s Thomson had succeeded only in getting grudging permission from the Admiralty to install one of his compasses, at his expense, on some suitable occasion yet to be determined and on a ship

yet to be selected, for the purposes of testing. He was becoming wily, though. Through his long association with cabling, his work with the Admiralty on other matters, and by his frequent sailing around Britain on the Lalla Rookh, Thomson was a familiar and cheery figure among naval officers around the country. He made his compass available informally to a number of captains, a strategy that Admiralty men regarded as a low trick and which may have hardened opposition. However, he began to win allies. Favorable reports on his compass trickled in, and a base of support grew among mid-level naval officers. The Thomson compass encountered problems, particularly a tendency to become unstable under rough conditions. Thomson refashioned the gimbals and the card suspension. Some Admiralty officials regarded this as a variety of cheating. Into the 1880s the Thomson compass made inroads in commercial shipping and was gaining a few crucial promoters in the Navy. Evans and Mayes and their Compass Department, however, remained steadfast in their determination to hold on to the now 40-year-old Admiralty Standard.

***

Energy, a word originally born of science, has emerged into everyday language. Entropy has come part way, used occasionally in nontechnical contexts with varying degrees of accuracy and appropriateness. Some words never make it out of the scientific lexicon: enthalpy, diamagnetism, quaternion. And some miscarry within science itself. In 1884 Thomson delivered a series of lectures at Johns Hopkins University in Baltimore, in the course of which he amused his audience with a novel and curious terminology: thlipsinomic, platythliptic, plagiotatic, cybotatic, euthythliptic, and more besides. No modern physics student will recognize these words or could guess what they refer to. A physicist today reading Thomson’s Baltimore lectures, as they became known, might well be able to follow the author sentence by sentence, equation by equation, but the purpose of the whole intricate exercise would seem opaque. Celebrated as they were in Thomson’s lifetime, the Baltimore lectures stand as an elaborate monument to a forgotten cause, like one of those architectural follies wealthy Victorians liked to put up in some bosky corner of the estate to surprise visitors.

Thomson first visited the United States in 1876, when he acted as a judge in the technical instrumentation section of the Centennial Exhibition in Philadelphia. Sailing across the Atlantic with Lady Thomson on the S.S. Russia, he had kept the crew and passengers entertained with constant experimenting on his compass and a new version of his sounding machine. Returning on the Scythia, he claimed to have found a previously unknown shoal somewhere in mid-Atlantic, where he plumbed a depth of only 68 fathoms at a point where the charts said 1,900 fathoms. Oceanographers have inexplicably failed to rediscover this shelf.

Among the dazzling array of inventions he saw in Philadelphia, Thomson particularly noticed an automatic telegraph receiver and an electric pen presented by the 29-year-old Thomas Edison, whose remarkable career was just beginning. Edison had started as a junior telegraph operator, teaching himself electricity and some engineering. Just as telegraphy was the first commercial technology to make use of the science of electricity, so the telegraph industry offered a route into technical careers for those, like Edison, who did not tread an academic path.

Thomson also saw a liquid compass by E. S. Ritchie of Boston, in which a card and needles floated on water in an enclosed vessel instead of being suspended in air on pivots and gimbals. The U.S. Navy had already adopted the liquid compass, a decision that was to prove wise in years to come. Ritchie sent one of his devices to the Compass Department in London, but Evans brushed off this innovation, all the more easily, no doubt, as it was not only radical but foreign. Thomson likewise showed no great enthusiasm for Ritchie’s compass. He had already designed his first dry-card compass and wasn’t about to be deflected from his purpose by an entirely different design.

Most astonishing, though, was Alexander Graham Bell’s telephone. Thomson heard “marvellously distinct” the words “to be or not to be” spoken through the device, along with a selection of items read from a local newspaper that were not so easy to make out. He brought back a pair of telephones from Philadelphia to show off at the British Association meeting in Glasgow later that year but had some difficulty with Bell’s primitive microphone and couldn’t get the apparatus to perform. (Edison’s “button” microphone of the following year, using compacted

carbon powder whose resistance varied with applied pressure, made the telephone into a far more practical instrument.)

After Philadelphia the Thomsons went on a whirlwind train tour taking in Niagara Falls, Toronto, Montreal, Boston, and Newport. At the BA meeting in Glasgow, Thomson spoke enthusiastically of “the originality, the inventiveness, the patient persevering thoroughness of the work, the appreciativeness, and the generous open-mindedness” that he had seen on display in Philadelphia. He noted sharply that in America “every good thing deserving a patent was patented” and told his audience that the “onerous” British patent system was “far behind America’s wisdom in this respect” and that if the British and European patent laws were not amended “America will speedily become the nursery of useful inventions for the world.” The New York Times saw fit to reproduce these glowing remarks.

Eight years later the British Association organized its first meeting abroad, in Montreal. Thomson was part of a large contingent from Britain, and he and several others went on to Philadelphia to attend a meeting of the American Association for the Advancement of Science, a counterpart to the BA founded in 1847. In both Montreal and Philadelphia the local newspapers splashed accounts of the visiting luminaries on their front pages. Science was the driving force of the age, and its achievements, most recently the telephone, seemed little short of miraculous. “To see such men is a privilege,” declared the Montreal Gazette of the city’s eminent visitors. “The meeting of the British Association … has been one of the happiest events in our history and one from which much and manifold good may be reasonably expected.” The paper devoted dutiful pages reporting to Montrealers the arcane discussions of the visiting savants.

Philadelphia, more cosmopolitan and confident, was not quite so overawed, but still the Inquirer ran a long account of the many famous men coming to town for the AAAS meeting, notable among them “Sir William Thomson, England’s great mathematician and electrician.” Thence Thomson went on to Baltimore, where in a story on events at Johns Hopkins the Sun announced that “the great event in the year’s work will probably be the lectures by Sir William Thomson … considered by many scientists second only to Newton.” Daniel Coit Gilman,

president of Hopkins, had written to Thomson in 1882 inviting him to deliver a series of lectures on whatever subject he cared to choose, telling him that this “would give a strong impulse to the study of Physics in this country.” Although Thomson, after his earlier visit, had praised the technical inventiveness of the Americans, academic science was in a rudimentary state. Johns Hopkins was at that time probably the nearest to a European research institution, with departments of science, graduate students, and noteworthy professors.

Gilman asked Thomson to start with a general talk to a large audience but emphasized that the point of his lecture series would be to introduce a select group to the most advanced topics and pressing questions. He passed on advice from Wolcott Gibbs, chemistry professor at Harvard, who believed that “the very best and most effective—most stimulating—course would be one on the obscure and difficult points in our modern physics. For instance on the difficulties we meet in the wave theory of light, in the atomic and molecular theory of matter, in electricity, as regards the want of any physical theory whatever.” Gibbs wanted “a really vigorous showing up of our shortcomings, especially if supported by new views such as Thomson could and would bring forward…. Every professor of physics in this country would want to hear such a course.”

Thomson settled on the wave theory of light as his theme. Maxwell’s theory, according to which light was a form of electromagnetic radiation, was by then two decades old but not yet widely understood or accepted. Heinrich Hertz’s laboratory demonstration of radio waves was four years in the future. Maxwell’s one undeniable success, in Thomson’s estimation, was the connection he found between the speed of light and the propagation of purely electric and magnetic phenomena in a vacuum. To modern thinking, this alone almost demonstrates the fundamental correctness of Maxwell’s theory. As far as Thomson was concerned, it was a pregnant quantitative prediction that the current evidence supported but by no means proved beyond doubt.

In particular, Thomson found Maxwell’s theory deficient because it had nothing to say on precisely what constituted electric and magnetic phenomena, on what light was, or on how these effects passed through a vacuum and interacted with matter. Maxwell proposed certain general concepts—electric and magnetic fields—and showed mathematically how

they related to each other. He showed that these interlinked fields could sustain oscillations that traveled at a fixed and finite speed. Defining the strength of electric and magnetic forces according to their respective inverse square laws are two constants known respectively as the permittivity and permeability of the vacuum. These two constants are linked in a simple way, Maxwell showed, to the predicted speed of electromagnetic radiation. The speed thus calculated was suggestively close to the speed of light, according to the best available data.

Thomson accepted the importance of the speed prediction but in other respects disliked what he saw as the abstract nature of Maxwell’s theory. How do electric and magnetic effects propagate through space, and what physical mechanism determines the permittivity and permeability? Surely what we call empty space must be a medium of some kind, if it is to support wave motion. What constitutes this medium and how does a wave motion manifest itself? Sound waves in air, every physicist knew, were pressure waves, the air bunching up and spreading out alternately. What, Thomson wanted to know, was the corresponding picture for electromagnetic waves? On these issues Maxwell was silent. He simply labeled the vacuum by certain parameters, characterized electromagnetic fields by certain mathematical functions, and proposed relationships among these things. It was a start, Thomson agreed, but it was not yet physics as he understood the term.

When one thought of light and electromagnetism and their behavior in matter, moreover, further difficulties and complications arose. To some extent, Maxwell could model various materials—insulators, conductors—by using appropriately adjusted values for the permittivity and permeability. But this merely glossed over the fundamental questions, Thomson believed. Why was one material a conductor and another an insulator? Some materials responded to a magnetic field by becoming magnetic themselves, in the same sense as the applied field. Others developed magnetism that opposed the applied field. Why? Again, Maxwell’s theory allowed these phenomena to be given mathematical labels, but that didn’t explain anything. Thomson wanted to know what went on in inside a material when an electromagnetic influence pervaded it. He wanted models that would explain and predict the behaviors that Maxwell’s theory merely accommodated and labeled.

Finally, there were phenomena that Maxwell’s theory failed to ad-

dress at all. Newton long ago had shown that pure white light could be split by a prism into an orderly rainbow of colors. In 1814 the German astronomer Joseph von Fraunhofer found that hundreds of dark lines crossed the spectrum of light from the sun, and in the mid-1800s Robert Bunsen and Gustav Kirchhoff showed how these characteristic lines, appearing at certain fixed wavelengths of light, indicated the presence of individual chemical elements. Thus was born the science of spectroscopic analysis.⁷ But what was the physical mechanism by which some substance snipped out a handful of little sections of incoming light, rather than absorbing evenly across the whole spectrum? Maxwell had no answer. Thomson couldn’t accept a theory of light with nothing to say on so elementary an issue.

He began his series of lectures on October 1. A reporter for the Baltimore Sun dropped by to see the celebrated scientist in action and wrote that “the lecturer is a man tall, though somewhat stooping, with kindly eyes, gray hair, and broad high forehead. He speaks easily, but has a habit of constantly twitching his hands while addressing an audience.” Thomson had turned 60 a few months earlier but was still a slender, lively man, the limp from his shortened leg exaggerating the impression of constant activity.

He intended his lecture series as an extended collegial seminar, engaging his audience of about 20 in discussions that led to consultation of books and papers, augmented by hasty overnight calculations. The course of one day’s discussion fed into the next day’s agenda. He had his topics in his head but prepared little for each session. The English physicist Lord Rayleigh (who was born into the upper crust as John William Strutt and acquired his title when his father died) attended about half of the Baltimore lectures and remarked to his son years later: “What an extraordinary performance that was! I often recognized that the morning’s lecture was founded on questions that had cropped up when we were talking at breakfast.” This spontaneous disorderliness pleased Thomson’s audience, Rayleigh thought, more than a set of carefully prepared talks would

have done. “They were very much impressed and he got some of them to do grinding long sums for him in the intervals,” he recalled. Writing at the time to his mother, though, Rayleigh gave a somewhat less sanguine view. Baltimore had been a success, he said, although the “lectures were quite in the usual Thomsonian style, a sort of thinking out loud in an enthusiastic incoherent manner.” J. J. Thomson (no relation), discoverer of the electron, remarked of William Thomson that “he has been known to lecture for an hour before reaching the subject of the lecture. It was only very rarely that he prepared either a speech or a lecture. There was, to the few who were already interested in the subject he was talking about, generally both charm and interest in these diversions.”

Ostensibly, Thomson talked in Baltimore of the wave theory of light. In the printed version of his lectures (they were stenographed and reproduced), he appears to talk at great and often mystifying length about waves in fluids and solids with various presumed characteristics, and even more enigmatically of the oscillations of imaginary mechanical constructions that he asked his listeners to ponder. There were apparently featureless spheres, inside which lay concealed, like Russian dolls, smaller spheres linked to the adjacent ones with springs (and zigzag springs, mind, not the usual spiral sort). There were geometric arrays of rigid rods, joined in such a way that they could rotate and pivot in a restricted fashion. There was a flywheel on an axle inside a sphere; but that wasn’t complicated enough, so Thomson proposed two flywheels inside the sphere, on a split axle that could pivot about its midpoint.

For one example Thomson devised an actual model, which became known as the “wiggler.” On a steel wire suspended from the ceiling half a dozen short wooden slats were attached, like the rungs of a ladder. At the ends of the slats weights were placed, with bigger weights on the higher slats. The whole array could be oscillated by a pendulum attached at the bottom. By varying the frequency of oscillation Thomson showed a great variety of motions of the wiggler, with some slats going one way and some the other, at different frequencies, and perhaps with one in the middle remaining stationary.

These bizarre toys, in one way or another, were supposed to represent molecules of matter, or more specifically the way molecules interacted with light. Spectroscopy made it clear that matter responded not

just to light in a general sense but to particular frequencies of light, both absorbing and emitting at these preferred points of the spectrum. Thomson asked his audience to imagine that a molecule must be some sort of machine with a complex array of internal vibrations and oscillations. Light could set the thing going (in which case a particular frequency would be absorbed, as light energy went into the molecule) and once it was vibrating the toy molecule could emit light again—perhaps at the same frequency but in general (as in the wiggler) at some other frequency that had been also set going.

Thomson also discussed at length the propagation of light through the unknown medium—the ether—that sustained electromagnetic oscillations. The problem was to get the right physical characteristics for the ether, so that light behaved as it was empirically known to behave. He told his audience of a favorite demonstration. In his Glasgow laboratory he had almost filled a glass jar with water, thrown in some corks (which naturally floated), poured in a two-inch layer of wax (Scottish shoemaker’s wax, to be precise), and then scattered a few bullets on top. This arrangement sat there, doing nothing at all as far as the eye could tell. But after six months, as Thomson fondly explained, the corks and the bullets had vanished from sight. After a year, the corks had emerged on top of the wax, having floated slowly upward, while the bullets had sunk through and dropped to the bottom of the jar. A familiar medium, he concluded, could seem solid by the hour, but fluid by the month or year. This was relevant in understanding light, because the ether, to sustain the extremely rapid vibrations of light waves, must in some sense be rigid—just as a chunk of hard metal, when struck, will ring at high pitch, whereas a more pliable block of wood would give a dull thud. On the other hand, the ether must also be forgiving and tenuous, for the simple reason that slow-moving solid objects (such as the earth) apparently passed freely through it, with no hindrance.

Therefore the ether, as an initial basis for contemplation, was something like a wax—hard on short timescales, soft with respect to slow changes. Of course no real wax truly mimicked the required physics of the ether. The point was that such things were possible, broadly speaking, and therefore conceivable. He mentioned Burgundy pitch and Trinidad pitch and Canada balsam as having properties interestingly different from

his shoemaker’s wax. He talked of glycerin. The ether was like a wax or a jelly, he insisted; it was a matter of coming up with the appropriate characteristics which, Thomson admitted, he had not yet been able to do.

In the printed version of his 20 Baltimore lectures, Thomson occupies more than 200 pages with seemingly endless calculations of the vibrations and oscillations of increasingly rococo arrangements of waxes and jellies, rods and springs, passive weights and gyrostats, in all combinations. It is Thomson’s scientific style at its acme. Take a handful of ingredients, all uncontroversial and with well-understood properties. Combine them, to discover what kinds of behavior they present and how the phenomena might relate to the passage of light, or the interaction of light with matter. If the modeled phenomena do not have the required range and complexity to simulate their empirical counterparts, then add something: More springs! Another flywheel! Another set of hinged, frictionless rods! Thomson displayed no doubt that the strategy was correct. His response to any difficulty was to add more bells and whistles. There was no hesitation, no going back.

The curious vocabulary—thlipsinomic, platythliptic, plagiotatic, and the like—came up as Thomson cataloged the entire range of physically allowable behavior of three-dimensional solids with density, compressibility, and viscosity. There is simple compression, as when a pastry cook squeezes a lump of dough into a ball. There is shear, as when the cook, with the heel of the hand, pushes dough across the board to create an elliptical disk. And there is torsion, or rotation, as when a rope of dough is twisted to make a spiral. Pastry dough, however, is what physicists call an isotropic material. Its properties do not have any directionality. By contrast there are layered materials, such as mica or graphite, which may slide easily when pressure is applied in one direction, but buckle or crack under a force applied perpendicularly.

In general, a substance that isn’t isotropic responds to an applied stress with a distortion at some angle to the applied force. To account for the full range of strains produced in a maximally nonisotropic solid, 21 independent coefficients are needed. Thomson’s thermodynamics and engineering colleague William Rankine (who had died in 1872 at the age of 52) had been a man of classical learning and had devised a set of 21 names for these coefficients, according to the geometrical relationship of

stress and strain they each denoted. Rankine, Thomson explained in one of his numerous Baltimore digressions, had a particular obsession with the way English had acquired an erroneous pronunciation of Latin and Greek words by absorbing them through French and campaigned quixotically for reform. He “was the last writer to speak of cinematics instead of kinematics,” Thomson said. “Cyboid is a very good word, but I do not know that there is any need of introducing it instead of Cubic…. Rankine was splendid in his vigour, and the grandeur of his Greek derivatives. Perhaps he over did it, but I do not like to call it an error.” In Rankine’s system one had to say Kikero instead of Cicero, which Thomson admitted was too much, and in the end he preferred a more conventional language. “Platytatic” and “platythliptic,” for example, became “sidelong normal” and “sidelong tangential.”

Whatever antique charm these words may once have possessed, the concepts they stand for have left no mark on modern physics. Thomson’s aim, described but not achieved in his Baltimore lectures, was to find an ether, characterized by the correct set of values of the 21 coefficients, that would support oscillatory light waves with precisely their observed properties and relation to electric and magnetic stresses. Maxwell’s theory, of course, did just that. But there was no physics, or not enough, in Maxwell, as Thomson saw it. He summarized his central objection: “I never satisfy myself until I can make a mechanical model of a thing. If I can make a mechanical model, I can understand it. As long as I cannot make a mechanical model all the way through, I cannot understand, and that is why I cannot get the electromagnetic theory…. That is why I take plain dynamics. I can get a model in plain dynamics, I cannot in electromagnetics.”

Thomson, in other words, did not literally think that space was filled with some version of Scottish shoemaker’s wax or Canada balsam or that matter was really composed of tiny spheres concealing springs and gyrostats. Unless, however, he could reproduce the mathematics of any physical phenomenon in terms of some directly appreciable mechanical model, he did not believe he had explained anything. Maxwell, early on, had used mechanical models similarly in order to arrive at the eventual form of his theory. He imagined, in one famous discussion, magnetic effects propagating through space analogously to the way rotation would

pass among spheres rolling against idler wheels interposed between them. But as he explored further Maxwell found that adhering to strict mechanical pictures limited his ability to understand the links between electric and magnetic phenomena, and he learned to rely on mathematical laws alone, even if they represented physical entities that had no immediately perceptible mechanical counterpart. In the end he abandoned mechanical pictures and, as one historian put it, presented his “Dynamical Theory of the Electromagnetic Field” in 1865 “stripped of the scaffolding by aid of which it had first been erected.”

Maxwell’s is the modern strategy. Pictures and analogies of all and any kind are frequently useful in drawing up ideas for new theoretical constructs, but in the end those constructs stand or fall by their internal mathematical consistency and their empirical usefulness. Thomson’s insistence that every theory must be reducible to a suitable arrangement of simple Newtonian ingredients limits the imagination far too much, and for no good reason. It is the ultimate expression of a “mechanical” view of the universe. The obviousness of this position was self-evident to Thomson, who neither would nor could provide deeper justification. On this subject he was the last holdout.

***

Thomson adjusted and modified his compass and other nautical instruments with just as much ingenuity and resourcefulness as he fiddled with ether models. For such technical inventions his strategy was sound. During the 1880s word of Thomson’s compass spread around the Royal Navy, and ships’ masters began to carry them surreptitiously, against official policy. One man in particular become a crucial and outspoken booster. John Arbuthnot Fisher, born in 1841 in Ceylon (now Sri Lanka), became a midshipman in the Royal Navy at the age of 13, having passed an entrance examination that consisted, his biographer reported, of “writing out the Lord’s Prayer, and jumping over a chair, naked, in the presence of the doctor; after which he was given a glass of sherry as evidence of his having become a naval officer.” By the age of 18 he had command of his own ship, on which he oversaw a technical advance, the firing of guns by electric impulses coming from crude batteries of zinc and copper plates immersed in vinegar.

Fisher was a blunt, outspoken man whose career teetered constantly

on the edge of insubordination but was carried off with sufficient brilliance that he ended up, years later, as First Lord of the Admiralty. He was the kind of officer loved, if feared, by his men, respected warily by his fellow officers, and barely tolerated by the Admiralty. He was a reformer and an enthusiast for scientific innovation. In his memoirs he raged against mindless official resistance to any kind of change. “We still have ancient Admirals who believe in bows and arrows,” he fumed. “Didn’t the Board of Admiralty issue a solemn Board Minute that wood floated and iron sank? So what a damnable thing to build iron ships!” The merest hints of change, he complained, were reliably opposed by “some Commander Knowall … Admiral Retrograde … and then some old ‘cup of tea’ writes to the Times … these carbonised cranks who wield the pen, actuated by the wrong kind of grey matter of their brain.”

For a few years he taught gunnery school in Portsmouth, drilling young seamen in the mastery of another new technology, the torpedo. He succeeded by force of personality as much as pedagogic skill. One student, asked to explain why π was equal to 3.14159, supposedly wrote that it was “the most suitable number Captain Fisher could think of.”

Given command of the Northampton, Fisher first encountered Thomson and his compass in 1879. In cold weather and wearing a thin overcoat, Thomson spent hours on deck adjusting his compass, while young officers sent to assist him came and went shivering. Fisher told him at some point to come in from the cold, but Thomson assured him: “No, thank you, I am quite warm. I’ve got several vests on.” He then explained to Fisher his theory, allegedly acquired from the Chinese, that many thin layers were better than a few thick ones. Thomson, like Fisher, was robust, indefatigable, always interested in new ideas, and impatient with the past. They became great friends and allies.

Two years later Fisher, still only 40, became captain of the Inflexible, the largest ship in the Royal Navy. Invited by Thomson to dine one evening at the Royal Society, he saw Joseph Swan’s new incandescent light⁸ and immediately decided he needed them for his ship. His con-

stant requests for modifications and improvements (he stirred up trouble by insisting on more toilets) caused him, so he said, to be “regarded by the Admiral Superintendent of the Dockyard as the Incarnation of Revolution.”

The electric lighting on the Inflexible ran at first on a 600-volt system, and on one occasion a sailor got a bad shock from touching a poorly insulated wire. Fisher asked Thomson, who happened to be on board, to take a look. “He diagnosed the matter as ‘a nasty little leak, but not likely to be dangerous to life’,” Fisher recalled. “Just then the cable slipped through his hand and the bare wire touched his finger. He leapt into the air, and his immediate second diagnosis was ‘Dangerous, very dangerous to life. I will mention this to the British Association.’” In fact, a man was electrocuted not long after in a similar accident. The potential was cut to 60 volts.

Thomson’s utter lack of embarrassment at changing his opinion so immediately impressed rather than irked Fisher as the sign of man, like himself, capable of adapting unhesitatingly to circumstances. They were both pragmatists. Fisher understood very well the intricate system of swinging a ship and obtaining correction tables, but inveighed against its complexity, its lack of practicality, and especially against the dim-witted bureaucracy that had caused it to survive long past its useful lifetime. As a young man Fisher had found his way to a new ship in the eastern Mediterranean by hitching a ride, or whatever the nautical expression may be, on a tramp steamer heading out from Italy. The captain of this old vessel had been plying about the Mediterranean his whole life and told Fisher, who was curious about the man’s informal navigational practices, that he generally got about successfully by knowing his “lamp-posts”—the light-houses—and by having his engineer tell him how many turns the engine had made, from which he could figure how far they’d sailed.

“Well,” Fisher asked, “what do you do about your compass? Are you sure it’s correct? In the Navy, you know, we’re constantly looking at the sun when it sets, and that’s an easy way of seeing that the compass is right.”

“Well,” the old captain explained, “what I does is this. I throws a cask overboard, and when it’s as far off as ever I can see it, I turns the ship round on her axis. I takes the bearing of the cask at every point of the

compass, divides by the total number of bearings, which gives me the average, and then I subtracts each point of the compass from it, and that’s what the compass is wrong on each point. But,” he concluded, “I seldom does it, because provided I make the lamp-post all right I think the compass is all right.”

In his way, as King Edward VII told him many years later, Fisher was a socialist. He loathed the British class system and the privileges accorded to the genteel members of society. “We fight God when our Social System dooms the brilliant clever child of the poor man to the same level as his father,” he wrote (he had been brought up by his maternal grandfather, who he said was swindled out of his painstakingly acquired means). The compass correction system, so far as Fisher saw it, was a microcosm of aristocratic elitism and conservatism.

Having become a champion for Thomson’s compass, Fisher took no small pleasure in battling the Compass Department at every opportunity. “It was an immense difficulty getting the Admiralty to adopt [Thomson’s] compass. I was reprimanded for having them on board. I always asked at a Court-Martial, no matter what the prisoner was being tried for, whether they had [Thomson’s] compass on board. It was only ridicule that got rid of the old Admiralty compass…. But what most scandalised the dear old Fossil who then presided over the Admiralty compass department was that I wanted to do away with the points of the compass and mark it into the three hundred and sixty degrees of the circle (you might as well have asked them to do away with salt beef and rum!) … the ‘Old Salts’ said at that time, ‘There he is again—the d—d Revolutionary!’”

The “dear old Fossil” in question was either William Mayes or his superior and mentor Captain Evans, who, like Fisher, had gone to sea as a midshipman at the age of 13 and worked his way to the top. Both men were Navy through and through and knew no other life. Evans’s devotion manifested itself in an unremitting determination to adhere to tradition and obey official regulations to the smallest of the small print. Even Captain W. E. May, a historian generally sympathetic to the Compass Department and somewhat hostile to Thomson, described Evans as “pig-headed and self-opinionated. Once in the Compass Branch he had his orders and he meant to stick to them.” Fisher was no doubt opinionated too, but he was flexible and adventurous; his commitment was to a

mythical naval history of heroic deeds and courageous individuals saving the day over the pedantic objections of desk-captains in thrall to their rulebooks.

Not only Fisher but other captains began to rely on the Thomson compass and press for its official adoption. In 1883 the Admiralty relented a little, approving the use of the compass though only in a subsidiary relation to the Admiralty Standard. Still, according to Fisher, Mayes was plotting against the good cause. One day in 1885 he had been talking at the Admiralty with a captain who complained that his Thomson compass had been so poorly located, with respect to the iron structure of his ship, that he could hardly use it. By chance Mayes appeared just at that moment, and Fisher roundly declared to his colleague: “I can state from long experience that Capt Mayes may be relied upon to use every exertion to place Sir Wm Thomson’s Compass in the worst possible position.” “The result of this speech,” Fisher later told Thomson, “was most gratifying—I am convinced that the proper way to treat Capt Mayes is to deliberately and calmly insult him.”

In the meantime Thomson used the patent courts to fight off various competitors, with a determination that was often more thorough than admirable. His own compass, with a couple of exceptions, was the result of his putting together a variety of ingredients from numerous sources. He particularly defended his compass card—aluminum ring, small needles suspended on threads, light but with long period—as his chief innovation, but other compass builders with slightly different layouts could claim, with some justice, that these ideas had been floating about for some time before Thomson put them all in one card. He initiated a number of legal challenges against competing designs and won all of them, in one case by appeal to the House of Lords after a lower court had gone against him. Thomson had financial resources and friends in high places; the justices of Great Britain had no expertise in deciding technical questions; and Thomson’s reputation and sometimes hectoring manner in court overcame the opposition. Some years later Thomson appeared as an expert witness in a patent case involving electric wiring systems. A clever barrister with some technical knowledge seized on a small error Thomson made in his testimony to push the case for his side. After he had come down from the stand, Thomson was told of his slip-up and

somehow blustered his way back on to the stand where he began to deliver an impromptu technical lecture. As J. J. Thomson recounted the anecdote, the barristers objected to the judge: “‘My Lord, what has this to do with the case?’ ‘I don’t know! I don’t know!’ said the judge, and Thomson went on.”

At the end of 1887 the Admiralty appointed a new superintendent of compasses, Captain Ettrick W. Creak F.R.S., a man both scientifically knowledgeable and forward looking. He did not doubt that the old Admiralty Standard Compass, now coming up to its 50th birthday, had been kept on long past its natural lifetime. But that did not make him eager to take on the Thomson compass. The U.S. Navy, back in the 1860s, had decided to go with liquid compasses of the kind that Ritchie exhibited in 1876 in Philadelphia. The trick had been to design a chamber in which a compass card could float stably on water, under conditions of constant pressure, and with this and other improvements accomplished the liquid compass had far fewer of the dynamical problems associated with a dry card balanced on a pivot.

The Royal Navy had recently commissioned a number of fast torpedo boats, and in these, when they moved at high speed, or in rough conditions, or with weaponry firing, both the old Admiralty Standard and the Thomson compass proved unstable and useless, while a simple liquid compass remained level. Creak argued for the adoption of a range of compasses, including an improved Admiralty Standard, the Thomson compass, and liquid compasses for torpedo boats and for gunnery positions on other ships.

The ponderous mass of the Admiralty Board, however, prodded relentlessly by Fisher and others, had by this time finally turned from its old course enough to embrace the Thomson compass, and having belatedly and inelegantly come around, their Lordships were not inclined even to consider another technology. Although Creak mustered evidence in favor of the superiority of the liquid compass in some circumstances, the Board decided, on November 19, 1889, to make the Thomson compass the sole official compass of the Royal Navy.

For Thomson this was the final victory and vindication. At the time of the decision, Thomson was staying in London with Fisher, and when he returned to Glasgow at the end of the month he told his sister Eliza-

beth that “much mean and underhand work has been brought to light.” Some 60 letters from ships’ captains concerning the Thomson compass had allegedly been squirreled away at the Compass Department, suppressed by Evans and Mayes, and it was Fisher’s irresistible force that had brought them into the open. A few of these letters contained critical remarks but fully 51 (so Elizabeth King wrote to her daughters) “spoke in terms of unbounded admiration and appreciation…. I believe this has been going on for years, and that Admiral Fisher has been instrumental in exposing the abuse…. Uncle William does not want it talked of.”

In his history of the Compass Department, however, A. E. Fanning tells a different story. The 51 letters existed but were old. When the board met to make its final decision, Creak (being an honest man, Fanning says) dug out these old testimonials and presented them along with 24 more recent reports, which described difficulties with the Thomson compass as well as the virtues of the liquid compass. “His compass … was excellent for many applications, but for the requirements of the Navy of the 1890s its introduction was a retrograde step,” Fanning concluded. It was, after all, more than 13 years since Thomson had first taken out a compass patent, and although he had made many modifications since then, the fundamental design remained the same.

In 1883 Creak, then an assistant at the Compass Department, had written to Thomson congratulating him on the award of the Copley Medal by the Royal Society. “I may single out amongst the many practical results of your researches the benefits you have conferred on Navigation. Foremost amongst Navigational instruments comes your compass, and your steady advocacy of that instrument against adverse forces has made me—perhaps one among many—long since review the position I had taken up and thank you for having made me think the matter over again increasingly to the advantage of your conclusions. You can review the position of your compass as regards the Navy with pride and satisfaction.” Years later, though, after he had retired, Creak confided to a friend: “When the Thomson compass was first introduced as Standard Compass on board I felt it my duty to try and make it a success. It was, however, in many respects the bete noire of my existence.” It was not until after Creak had retired that his successor was able finally to introduce liquid compasses into the Royal Navy, Thomson having more or less retired from

the scene by this time. Thomson’s compass, though, could still be found on merchant vessels and other ships well into the middle of the 20th century.

Scientific biographers of Thomson, if they have paid any attention at all to his compass innovations, have generally taken the matter to be a sorry saga of dim-witted naval administrators resisting marvelous innovations from a superlative scientific mind. Writers sympathetic to the Navy, on the other hand, portray Thomson as a man of undoubted talent and enthusiasm, with some genuine knowledge of the sea, who managed to parlay a handful of modest ideas in compass design into a commercial monopoly for his own manufacturing concern, using his reputation as a bludgeon in the law courts to beat down even small claims of originality from others, and persuading the Admiralty and the law to overlook both the deficiencies of his own design and the virtues of his competitors’.

The truth, inevitably, seems to lie somewhere between these extremes.