The Golden Age Of The Iron Bridge

Iron was the miracle material of the nineteenth century—abundant, cheap, and extremely useful. It was perfect for jobs requiring great strength in proportion to weight: cylinders for pumps and steam engines; boats and barges for canals; beams and columns for mill buildings; and, eventually, bridges. Several thousand iron bridges were built in America between 1840 (when iron began to replace wood and stone) and 1880 (when it was in turn being superseded by steel); some six dozen survive.

These bridges used cast iron for compression members and wrought iron for tension members. Steel at the time was prohibitively expensive; wrought iron cost twice as much as cast iron but resisted tension (stretching) so much better that it was worth the expense for certain parts of a bridge. Wood was still often the cheapest material for much of the century, but it grew increasingly obsolete. Iron was the modern wonder—strong, affordable, mass-producible, portable, fire-resistant, and capable of being shaped into the loveliest designs.

Most of the iron bridges that still survive are in the eastern United States, chiefly in New York, Pennsylvania, and Ohio, with a few in New England and a few more in the Midwest and mountain states. Of all the basic American bridge types—including rustic covered wooden, stalwart stone or concrete arch, foursquare steel girder, and graceful suspension—the rarest and least appreciated is the cast- and wrought-iron truss. Yet in some ways it is the most technologically significant. Every molding, pin, nut, and bolt, the way the diagonals, verticals, top chords, and eyebars are fastened together, all help these gravity-defying structures resist the forces of nature and mankind.

You can feel a truss bridge work by holding its members as an automobile crosses. The deck gives way under the weight of the load but regains its original level once the car has passed. The wrought-iron diagonals stretch, and the upper chord deflects slightly; then they too return to their original configurations. The bridge is an organism, and it functions as it does because of the elastic properties of its materials, the quality of its workmanship, and the way its parts are assembled. When you consider how many times one of these bridges has been asked to flex over a hundred years, you can begin to understand its remarkable qualities.

Unfortunately America’s iron bridges have been almost universally disregarded and forgotten, even by preservationists and engineers. They exemplify fundamental American values of craft, entrepreneurialism, and creativity. They helped Americans cross thousands of streams and rivers, reach new markets, and create new businesses as the frontier moved west. Hundreds of patents on iron trusses were granted for them; while many went to trained engineers, others were awarded to crafters, millwrights, and mechanics. These unschooled “apple-tree engineers” recognized a need and sought engineering solutions that proved to be practical, if sometimes bizarre. If they are not saved now, these irreplaceable artifacts of the American landscape, both rural and urban, are threatened with extinction.

A truss is simply an interconnected framework of beams that holds something up. The beams are usually arranged in a repeated triangular pattern, since a triangle cannot be distorted by stress. In a truss bridge two long, usually straight members, known as chords, form the top and bottom; they are connected by a web of vertical posts and diagonals. The bridge is supported at the ends by abutments and sometimes in the middle by piers. A properly designed and built truss will distribute stresses throughout its structure, allowing the bridge to safely support its own weight, the weight of vehicles crossing it, and wind loads. The truss does not support the roadway from above, like a suspension bridge, or from below, like an arch bridge; rather, it makes the roadway stiffer and stronger, helping it hold together against the various loads it encounters.

Ancient Roman rules of truss design were revived during the sixteenth century by Andrea Palladio, the Italian architect. After Palladio the concept was largely forgotten for nearly two hundred years until two Swiss carpenters, the Grubenmann brothers, Johann and Hans, built several covered wooden truss bridges. But although Europeans pioneered the truss bridge, America’s wide-open spaces and rugged terrain made it inevitable that the technique would see its greatest development here.

The growing, mobile population in America in the early nineteenth century needed transportation networks—first canals and turnpikes, then railroads—to settle and serve the vast, uncharted land. No country in the world needed good, permanent bridges more. The young nation was coursed by hundreds of rivers, thousands of creeks, and millions of acres of swamps. Physical unification was impossible until engineers and builders overcame the inconvenient intervention of water.

America’s truss lineage begins in the early nineteenth century with the Burr, Town, Long, and Howe trusses, which were made primarily of wood, or wood with some iron parts. It continues with the Pratt and Warren trusses, which were designed and first built in wood but later adapted for iron. This evolutionary process was essentially empirical, with the sizes of members determined from the results of previous failures or from study models. Then in 1841, toward the end of the empirical period, Squire Whipple, the father of iron bridge building in America, introduced his bowstring arch. It was the first successful all-metal truss bridge, and its design was guided by scientific principles.

One way to trace the development of bridge design is to look at patents. The first patented bridge system widely used in the United States was introduced by Theodore Burr in 1806. The Burr arch-truss combined an arch and a parallelchord truss, with the two elements working together: the arch carried the load, and the truss served as a stiffener. This system was used extensively on highways and railroads until mid-century. The next major step was Ithiel Town’s lattice truss, patented in 1820, which had a number of advantages for use in the wilderness. It could be built of plank, usually three or four inches thick; intricate joints were not required; all connections could be made of round oak pins; the chord and web members were all formed from timbers of the same size, four-by-tens, which were readily available in convenient lengths; and the piers and abutments could be light because they were freed from horizontal pressures, which would have come from using an arch.

Town’s design was the first straight truss bridge, with no arch, and its ease of erection made it a cheap and popular choice for canal aqueducts, turnpikes, and railroad bridges. Town accumulated quite a fortune by collecting royalties of a dollar a foot on bridges using his design. As early as 1831 he suggested that his truss might be made from iron, but no builder tried it until 1859.

After Town, the next successful patents (there were many unsuccessful ones) for trussing systems were granted to Col. Stephen Harriman Long. Long was an Army topographical engineer working for the Baltimore & Ohio Railroad who received four patents between 1830 and 1839. Like Whipple after him, Long based his trusses on scientific principles, and they were notable for their simplicity and ease of assembly.

The first important move away from wooden construction came in 1840, when William Howe (an uncle of Elias Howe of sewing-machine fame) patented the Howe truss. In profile it looked very much like the Long truss, but Howe specified vertical wrought-iron tension rods as well as heavy wooden diagonal compression members. This foreshadowed another combination of two materials—cast and wrought iron—ten years down the road. The Howe truss may be the closest that wooden-bridge design ever came to perfection. For simplicity of construction, rapidity of erection, and ease of replacing parts, it stands without rival.

In 1841 Squire Whipple patented the iron bowstring arch, and six years later he used it to illustrate America’s first text (and possibly the world’s first) on scientific truss-bridge design. Whipple’s major breakthrough was analyzing truss members as a system of forces in equilibrium. Thomas Pratt, an engineer, and his father, Caleb, an architect, developed and patented a truss in 1844 that was the reverse of Howe’s, with vertical wooden compression members and diagonal iron tension rods. The Pratt truss did not enjoy the popularity of the Howe at first because it was less rigid and more expensive, since it used more iron. As the cost of iron decreased, however, the price difference decreased as well, making the Pratt more attractive because of its simpler connections and more logical distribution of stresses.

The final important truss configuration was the Warren, or triangular, truss. It was an improvement on a design patented in France in 1838 by a Belgian engineer named Neuville. Though these were the primary patents, many surviving bridges have ingenious and diverse designs that weren’t patented. They show peculiarities of having grown out of wooden-bridge design, idiosyncrasies that were discarded one by one for reasons of complexity, inadaptability, and the emergence of better solutions. The trusses that survived the wood-to-iron transition and endured into the twentieth century—the Pratt and the Warren—were simple skeletal forms that adapted well to iron and eventually to steel.

Bridge engineering evolved differently here from the way it did in Europe. In Britain the crown represented private enterprise, and Parliament had to approve all public works, while in France financing and design emanated centrally from Paris. In America engineering came to be dominated by entrepreneurs and craftsmen, largely self-taught at first and interested in doing rather than studying.

The first decades of the nineteenth century were the era of the carpenter engineer. Timothy Palmer and Theodore Burr (both carpenters), Lewis Wernwag (a mechanic), and Ithiel Town (an architect) continued the wooden-truss methods of Palladio and the Grubenmann brothers with forms based on intuition, pragmatic rules of thumb, and knowledge passed from master to apprentice. Their objectives were simplicity, easy construction, and improved details, particularly the connections where parts came together. Large eastern rivers like the Connecticut, the Hudson, the Delaware, and the Potomac were bridged with wooden trusses of unprecedented length that astounded visiting European engineers.

As the country expanded, the need to cross vast distances interrupted by high mountains and wide rivers combined with a general shortage of capital to ordain the construction of cheap, rapidly built, temporary bridges. Settlers hoped to make their roads and spans more permanent later. European countries, smaller, richer, and long settled, were meanwhile building monumental structures for the ages.

The first engineering students in America were instructed at the U.S. Military Academy—West Point—which was established in 1802. Under the direction of Sylvanus Thayer, appointed superintendent in 1817, engineering education at the academy was modeled after that at France’s Ecole Polytechnique, which had been founded in 1794. Before starting the job, Thayer visited Europe to study military schools, armies, and fortifications, and his choice for professor of civil engineering was Claudius Crozet, a graduate of the Ecole Polytechnique. Academy-trained engineers knew theory and applied it, but there were not enough of them to address the needs of a fast-developing country. Consequently the crafts-man-millwright-founder, who learned his skills by apprenticeship, built at least as many bridges as the educated engineer in the early part of the century. And he had to be energetic, self-reliant, and ingenious, a jack-of-all-trades who could solve each problem anew with limited resources, little time, and few precedents. Bridging small streams was part of the pioneer’s life.

The world’s first iron bridge was cast in 1778 at England’s Coalbrookdale Ironworks and erected the following year over the River Severn, to demonstrate the versatility of cast iron. Designed by an architect, Thomas Farnolls Pritchard, it spanned one hundred feet on five cast-iron ribs. Today, as when it was built, it is the centerpiece of Ironbridge Gorge, one of the world’s great open-air museums and the destination of bridge aficionados worldwide.

The great pamphleteer of the American Revolution, Thomas Paine, was probably familiar with that bridge, and he made himself a champion of iron bridges as well as of liberty. Paine realized that stone and wood, though excellent in compression, were poor in tension—or in the case of wood, hard to splice in a good tension connection. In 1786 he designed a four-hundred-foot cast- and wrought-iron arch bridge with thirteenarched ribs—one for each state in the new union.

Today some skeptical European historians doubt that a radical political philosopher untrained in the building arts could possibly have conceived such a workable plan and suggest that he may have stolen the idea from a model he saw in Paris in 1781. Whatever its origin, Paine’s design, after being favorably reviewed by the French Academy of Sciences, was awarded a British patent in 1788. With the help of Walker Brothers, a family of iron founders in Rotherham, England, he built a prototype and displayed it for several weeks in front of the Stingo Pub in Paddington, near London.

Paine’s campaign for iron bridges ultimately came to naught, at least as far as America was concerned, and the material was not seriously considered in the United States until the 1830s, when Town and Long suggested that their wooden-bridge patents could be constructed in iron. In 1833 August Canfield, a West Point graduate working in Paterson, New Jersey, patented America’s first bridge designed specifically to be made from iron, but it was not practical and was never built. This country’s first all-iron bridge was the Dunlap Creek Bridge, completed in 1839, which still stands (see box on page 17). Erected on the Cumberland Road in Brownsville, Pennsylvania, it was an evolutionary anomaly. Instead of a truss, Richard Delafield built a heavier, more expensive cast-iron arch bridge. Presumably his West Point education made him aware of the favored style of construction in Britain and France. Though he claimed his idea was original, the concept is similar to the Pont du Carrousel over the Seine in Paris, built in 1834.

After Dunlap Creek inaugurated the use of iron, two bridges spanning New York State’s Erie Canal marked its earliest use in truss bridges. In 1840 Earl Trumbull built the first iron truss bridge in America, a span of seventy-seven feet at Frankfort, New York. Months later at Utica, Squire Whipple built his famous bowstring arch (so called because the shape of the top chord and the deck resembled a bow used in archery). Whipple was born in 1804 to a farmer in Hardwick, Massachusetts, and studied engineering at Union College, graduating in 1830. After first working as a rodman and leveler on the Baltimore & Ohio Railroad, he became a surveyor on the Erie Canal, and when engineering work was not available, he invented and manufactured engineering and surveying instruments. In 1841 he patented his bowstring arch truss; he later started one of the country’s earliest bridge companies and built hundreds of iron bridges for canal, rail, and highwav use.

In 1846 Whipple published a pamphlet called An Essay on Bridge Building , which the next year he expanded to book length—America’s, if not the world’s, first book on mathematical truss analysis. The volume was ignored for years, but it marked the beginning of rational bridge design in America, which would eventually raise the discipline from a craft to a profession.

During the 1840s iron bridge building in America made impressive strides. In 1845 Richard Osborne, a London-born Irish-emigrant engineer working for the Philadelphia & Reading Railroad, built the country’s first all-iron railroad bridge. It was known as the Manayunk Bridge, after the Philadelphia neighborhood in which it was built, and consisted of three lines of Howe trussing reaching thirty-four feet across a small stream. A single line of that trussing was donated to the Smithsonian Institution in 1911; today it serves as part of the exhibit of the John Bull , an early steam locomotive.

A period of rapid experimentation with the new material ensued. In 1847 Frederick Harbach built an iron Howe truss on the Western Railroad (later the Boston & Albany) near Pittsfield, Massachusetts. That same year James Milholland built a bridge for the Baltimore & Susquehanna Railroad at Bolton Station, Maryland, using girders—the first to do so in America and a design far advanced for its time. Iron girders would become a common feature of railroad bridges twenty years later.

These early designs were officially embraced by Benjamin Henry Latrobe, Jr., chief engineer of the Baltimore & Ohio. His line became the first railroad to adopt iron as standard for its bridges and moved into the forefront of advances in building bigger, stronger, longer bridges. Two engineers working with Latrobe, Wendel Bollman and Albert Fink, developed all-metal trusses of cast and wrought iron that could support more than one ton per lineal foot.

Bollman, a Baltimore pharmacist who switched to railroad work, ultimately became foreman of bridges and then master of the road for the B&O, responsible for all structures along the right-of-way. In 1858 he returned to private business, and he founded his own bridge company after the Civil War.

Albert Fink immigrated to the United States a year after graduating from the Polytechnic School in Darmstadt, Germany, in 1848. After Bollman refused him a job as a draftsman on the B&O, Latrobe hired him. Fink eventually designed and patented a truss of his own and became Latrobe’s principal assistant for part of the road. He went on to be vice president and general superintendent of the Louisville & Nashville Railroad.

Bollman’s first major span, erected in 1851 over the Potomac at Harpers Ferry, was 124 feet long. A year later Fink put up three trusses of 205 feet each, 43 feet above the Monongahela River near Fairmont, (West) Virginia. Both Fink’s and Bollman’s bridges were suspension trusses, with chords and posts of cast iron and diagonal tension members of wrought iron. The tension bars had eyes at their ends that were pierced by finely turned forged pins, instead of being held in place with cast-iron trunnion blocks. This freedom of rotation meant that the trussing system could be analyzed using simple mathematical formulas. Use of the Bollman and Fink all-iron pin-connected trusses by the B&O revolutionized railroad-bridge design. Soon after, the Pennsylvania Railroad began building iron Pratt trusses stiffened with castiron arches, a design that served as that railroad’s standard from 1851 to 1861.

After mid-century, trained engineers like Whipple, Osborne, and Fink assumed significantly greater roles in bridge design. The Whipple truss, with end posts inclined instead of straight up and diagonal tension members crossing two panels instead of one, was the first of a whole new generation of metal trusses. In 1853 Squire Whipple built a 146-foot span of this configuration for the Rensselaer & Saratoga Railroad near Troy, New York. He called it Whipple’s Trapezoidal Truss. In 1859 John W. Murphy built a 163-foot span over the Morris Canal at Phillipsburg, New Jersey, for the Lehigh Valley Railroad, that was patterned on the Whipple configuration but pin-connected throughout. It became known, to Squire Whipple’s chagrin, as the Murphy-Whipple, and it became the standard truss type for longspan bridges for the rest of the century.

These trusses were adequate for Eastern terrain, but as railroads expanded westward, they encountered a formidable barrier: the Ohio River. It had been bridged, of course, but as trains got heavier, much sturdier bridges were needed. Spanning the Ohio now demanded major improvements in scientific design and a deeper understanding of the way materials such as cast and wrought iron behaved under stress. Engineers developed more realistic mathematical models in which the weight of a train was concentrated at the engine and the axles, instead of being spread uniformly along the length of the bridge.

Wheel-load analysis was first proposed in 1862 by Charles Hinton, a bridge engineer for the New York Central Railroad, and it quickly became standard American railroad practice. In addition, the Pennsylvania Railroad introduced testing machines to help engineers understand and measure material behavior. In 1863 William Sellers, a Philadelphia machine builder, developed a machine with five hundred tons of capacity, capable of testing full-size structural members. This enabled Jacob H. Linville, chief engineer of the Keystone Bridge Company, to design and build truss spans so long they could bound the Ohio River.

The railroads had reached the Ohio on the eve of the Civil War; crossing it, however, still required a trip on a ferryboat. Then wartime demands pushed the technology of metal truss-bridge construction to near-perfection. By the end of the war, railroads were the most powerful economic and political force in the United States. They were ready to address the broad and meandering Ohio and Mississippi—an extensive, wide, fast-moving, and deep river system that separated the Eastern United States from the West and the North from the South.

To cross these rivers, engineers developed new, ever-longer-span trusses. The Steubenville Bridge, in Ohio, with a channel span of 320 feet and trusses 28 feet deep, was the first. Linville designed it for the Pittsburgh, Cincinnati, Chicago & St. Louis Railroad, using Whipple’s Trapezoidal Truss as his pattern. Its completion in 1864 began the era of long-span truss-bridge design in America. Albert Fink followed with a hybrid Whipple-Pratt bridge over the same river at Louisville in 1870, with main spans of 360 and 390 feet. The Whippie and Fink types both grew popular because they could cover long distances but had short panel lengths for even load distributions. The longest Fink-truss span ever built was only 306 feet, but the Murphy-Whipple, with its greater rigidity, reached an unprecedented 518 feet.

The most enduring achievement of the postwar period, however, was the bridging of the mighty Mississippi below its confluence with the Missouri. This astounding feat was accomplished in 1874 by James Buchanan Eads, another nineteenth-century Renaissance man who, like Paine and Bollman, was not a trained engineer. (See “Eads and the Navy of the Mississippi,” Invention & Technology , Spring 1994.) The Eads Bridge, the oldest and most graceful of the Mississippi River bridges, shattered all engineering precedents with its doubledeck design and its three ribbed steel arches that each spanned more than five hundred feet. The arches were cantilevered and supported by cables as they were being built, eliminating the need for falsework, which would have been impossible on the broad, deep, fast-flowing Mississippi. Pneumatic caissons (submerged, floorless cylinders pumped with air), rather than coffer dams built from the surface, were another innovation. And the Eads Bridge marked the first major use of steel in bridge construction.

The triumph of spanning the Mississippi in 1874 was offset in 1876 by the Ashtabula Bridge disaster. On a cold, miserable, snowy Ohio night a few days after Christmas that year, seventysix people perished when an 1865 cast- and wrought-iron bridge collapsed, plunging a passenger train into the Ashtabula River. Many railroad bridges had collapsed before, but never with so great a loss of life. The accident shook the engineering profession and weakened the railroad industry’s confidence in its bridges. The ensuing investigation and report by the American Society of Civil Engineers condemned combination cast- and wrought-iron bridges in favor of all-wrought-iron construction, riveted rather than pin-connected.

Three years later, in December 1879, the British railroad industry suffered a similar catastrophe when a combination bridge failed under a passing train. Thirteen spans of the two-mile-long viaduct across the Firth of Tay at Dundee, Scotland, collapsed, giving British engineers renewed respect for the forces of nature. A court of inquiry determined that the failure had been caused by ignorance of metallurgy, uneven manufacturing techniques, defective castings, and instability under wind loads.

In the wake of these collapses, bridges were examined in both countries, and many instances of defective design and construction turned up. Floor systems were too frail for the new, heavier engines. Broken castings illustrated the unreliability of cast iron even when used just for joint blocks (the pieces used to join one member to another). Connection fittings as well as the cross-sectional areas of members—the details upon which the capacity of a bridge depends—were found inadequate. Clearly, bridge design had a long way to go.

The number of bridge-fabricating companies had been exploding during the second half of the nineteenth century, nearly doubling every decade. In 1850 only 15 such companies were operating in the United States; on April 13, 1900, there were more than 190, and on that day J. P. Morgan consolidated 24 of them into the American Bridge Company.

One firm not absorbed by Morgan was the Phoenix Bridge Company, located about an hour out of Philadelphia on the Reading Railroad. It was a bustling enterprise established in 1790 where you could follow the building of a bridge from iron ore to finished product. At the machine shop wrought iron came from the puddling furnace to be rolled, planed, drilled, and riveted into the Phoenix column, a proprietary section patented by one of the owners, Samuel J. Reeves, in 1862. Everything was made on the most modern machines, allowing great accuracy and precise duplication of parts.

Samuel Reeves also developed the Phoenix bridge (yet another Murphy-Whipple configuration), which contributed to the demise of the composite cast- and wrought-iron structure. It used the Phoenix column, made entirely of wrought iron, and was so lightweight and strong that it rendered the price advantage of cast iron immaterial.

Before shipping a bridge to the spot where it would be erected, mechanics at Phoenix assembled the parts on the shop floor to ensure proper length and fit. They were marked so that workmen in the field could quickly reassemble them. Field erection required scaffolding—a falsework of wood capable of supporting the weight of the bridge until it was finished. If the site was a deep gorge, the falsework was often as elaborate and expensive as the finished bridge. The technique of prefabricating and preassembling all the parts to guarantee an exact fit and then securing the members with pins became known as the “American plan.”

Here, as in other aspects of bridge design and construetion, American and European practices diverged. In Europe, where bridges were riveted together, erection time was longer and the costs were consequently higher. A riveted Town lattice truss of 160-foot span took ten to twelve days to build; a Phoenix truss of similar span could actually go up in eight or nine hours. Eventually American engineers would opt for slower, costlier all-riveted construction, but the American plan prevailed in this country until the end of the century.

Partly because of this, American bridge manufacturers had the lowest prices and did the fastest work in the world and competed successfully in world markets. Despite the Ashtabula disaster, American bridge shops were by the 187Os as high tech as anything anywhere. They even employed the latest in corporate organization and structure. Morgan’s American Bridge Company had a financial department and auditor, a purchasing department, an executive and secretariat, and engineering, operating, contracting, and mechanical departments—the nucleus of a modern industrial corporation.

In the last quarter of the century, the ever-mounting demands of the railroads transformed American bridge engineering. Until 1876 bridge design and construction were almost exclusively controlled by companies that promoted one particular style—for instance, the National Bridge Company of Boston’s Parker truss or the Watson Machine Company of Paterson, New Jersey’s Post truss. After the Ashtabula disaster, railroads started holding design competitions based on specifications and performance standards instead of relying on off-the-shelf designs or in-house bridge building. A bridge over the Ohio River, made in 1876 for the Cincinnati Southern Railroad, was the first large bridge project awarded on the basis of a competition. The Keystone Bridge Company of Pittsburgh was the winner.

Written specifications were something new. The first set adopted by an American railroad was drawn up in 1871 by George Morison, an engineer for the Erie Railroad. (See “The Master Builder,” Invention & Technology , Fall 1986.) Morison went further than just setting forth specifications; he required successful bidders to submit stress diagrams, calculations, and plans before starting work, and later he personally inspected materials and workmanship in the field. Theodore Cooper wrote up specs in 1878—improving on ones written by Reeves and Morison—that became the basic form of instructions most widely used by American railroads for the rest of the century.

With the rise of design competitions, many top engineers left railroad engineering departments to enter private practice. By 1880 bridge consultants had become the first specialists within the general field of civil engineering. Bridges were the most highly developed kind of structure type, and consulting bridge engineers the most skilled and respected practitioners. They had come a very long way indeed from the frontier craftsmen who had dominated bridge building half a century before. Theodore Cooper characterized the successful bridge engineer of 1880 as not just a mere calculator of stresses but someone with a full knowledge of the practical and theoretical elements of design, manufacturing, and erection, as well as an “instinct of design.”

Today we need an equally sensitive and sophisticated specialist, capable and experienced, to preserve the outstanding bridges of the past. He or she must be familiar not only with contemporary computer modeling and structural analysis, but also with nineteenth-century building materials, and the theories, formulas, and rules of thumb used by Burr, Town, Wernwag, Whipple, Linville, Morison, and the hundreds of engineers and draftsmen who cranked out designs by the mile during the nineteenth century.

Few people willfully destroy historic bridges; we usually lose them through ignorance of their importance. To preserve the historic bridges of the United States, we need the cooperation of people in transportation and engineering; preservationists, architects, and historians working alone will only partially succeed. Bridges are engineered structures, and they require the ingenuity of engineers to survive. Above all, we need an informed public that recognizes the significance of historic bridges and insists that the best ones be saved—not only for practical reasons, but for the enrichment of posterity.