SERIAL PUBLICATIONS OF THE SMITHSONIAN INSTITUTION The emphasis upon publications as a means of diffusing knowledge was expressed by the first Secretary of the Smithsonian Institution. In his formal plan for the Insti- tution, Joseph Henry articulated a program that included the following statement: "It is proposed to publish a series of reports, giving an account of the new discoveries in science, and of the changes made from year to year in all branches of knowledge." This keynote of basic research has been adhered to over the years in the issuance of thousands of titles in serial publications under the Smithsonian imprint, com- mencing with Smithsonian Contributions to Knowledge in 1848 and continuing with the following active series: Smithsonian Annals of Flight Smithsonian Contributions to Anthropology Smithsonian Contributions to Astrophysics Smithsonian Contributions to Botany Smithsonian Contributions to the Earth Sciences Smithsonian Contributions to Paleobiology Smithsonian Contributions to Zoology Smithsonian Studies in History and Technology In these series, the Institution publishes original articles and monographs dealing with the research and collections of its several museums and offices and of profes- sional colleagues at other institutions of learning. These papers report newly acquired facts, synoptic interpretations of data, or original theory in specialized fields. These publications are distributed by subscription to libraries, laboratories, and other in- terested institutions and specialists throughout the world. Individual copies may be obtained from the Smithsonian Institution Press as long as stocks are available. S. DILLON RIPLEY Secretary Smithsonian Institution SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY • NUMBER 10 ROEBLING'S DELAWARE & HUDSON CANAL AQUEDUCTS Robert M. Vogel SMITHSONIAN INSTITUTION PRESS CITY OF WASHINGTON • 1971 FIGURE 1.—The Delaware Aqueduct, 1970, in the early morning river mists. (Photograph by author.] COVER: The Delaware Aqueduct, 1969. (Photograph by David Plowden.) ROEBLING'S DELAWARE & HUDSON CANAL AQUEDUCTS The nineteenth-century American civil engineer, John A. Roebling, is best remembered for his crowning work, Brooklyn Bridge, built to his design by his son, Washington, following the elder Roebling's death in 1869. Although an engineering monument of the highest order, Brooklyn Bridge must—if historical justice is to be done—share its notoriety with a small, relatively obscure sus- pension bridge that was Roebling's second work, and is his earlist still standing. Moreover, in all likelihood, the Delaware Aqueduct is the oldest existing Ameri- can suspension bridge and may well be the oldest existing suspension bridge in the world (that retains its original principal elements). The sole survivor and largest of four suspension aqueducts erected by Roebling between 1847 and 1850 to carry the Delaware & Hudson Canal over rivers, the Delaware Aque- duct stands today only because of its strategic location. Following abandonment of the canal in 1898, the structure was converted to a private highway bridge, which function it continues to serve, spanning the Delaware above Port Jervis. THE AUTHOR: Robert M. Vogel is Curator of the Division of Mechanical and Civil Engineering in the Smithsonian Institution's National Museum of History and Technology. The Delaware & Hudson Canal Unlike the Erie and most other American barge canals, the Delaware & Hudson Canal, opened in 1829, was built as an essentially one-way route to transport a single commodity — anthracite coal — rather than general freight in two directions.1 It was 1 The canal company did, of course, avail themselves of the opportunity to carry whatever along-the-line general freight presented itself, and this business, while far less significant than coal haulage, was a worthwhile account. For example, in the year 1846 the goods transported were: coal, 318,000 tons; general merchandise, 28,000 tons; and lumber, 10 million board feet. (Annual Report of the Board of Managers of the Delaware & Hudson Canal Company for the Year 1846. New York, 30 March 1847). projected by Maurice and William Wurts as a means of exploiting their great coalfields in northeastern Penn- sylvania, a canal at that time being the only feasible way of getting the bulk coal to the seaboard. As New York was potentially the most profitable market area, the canal was planned to strike for the Hudson River, down which the coal could be readily transported to the city. Charters were granted to the Wurts' by the Pennsylvania and New York legislatures to improve the navigation of the Lackawaxen River—reaching practi- cally into the Lackawanna coalfields at Honesdale and at its mouth joining the Delaware—and to build a line of water communication between the Delaware and Hudson rivers. The Delaware & Hudson Canal Company was 1 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY 'Arrran fcrosir ] KlMSSTDMOi CPff K LOCKS/ T ' ' HIGH FALLS, I FbuGHK-ffpaie- n J* /[LLfriVILLf 2/ \ /reryoueoMt I7AYW FAEVIFH 4r Gem'sry/P/P llllllllll • -...-.. locomrr/v**'*' — — — — — f***S—<^»ll«f T II. 0/&»-&tUJ*0*VI3 t \ ShtphtrtiCra* CABBONDALE JjpBVf^ 4%, ARCMBALD, Q OLYPMAMT^ "J*feTTli % VALLEY JCT | \« LHONESMLE LHAWLEY/ /ICtfTICfLLO * 'tlAMAKATiriG CuDDEBACKMILLE; \ jfjrtootfr IGo&r&y 1* i ?***£""" *M£ VVl* Fbreeson PL YMOUTH Thrrsron JV/Ltff-3 BAPBf- »■ a 1">M ** -" fffnrortri Pleefionr I flrrHAriKfri \tt*o* gjjjgg& FIGURE 2.—The Delaware & Hudson Canal and Railroad System, 1866. (From A Century of Progress: History of the Delaware & Hudson Company.) formed and in the spring of 1823 contracted with Benjamin Wright to survey and locate a suitable route. At the time, Wright was still serving as chief engineer of the Erie Canal. He was instructed to select a line from tidewater on the Hudson at Rondout (near Kingston), up the valleys of the Rondout, Nev- ersink, Delaware, and Lackawaxen rivers to the coal- fields. The total distance was 108 miles with a lockage of 1,086 feet.2 Construction began in 1825—the year of the Erie's opening—with Wright acting as chief en- gineer and the later renowned John B. Jervis as his as- sistant. The entire canal was opened for business in October 1829. Seven thousand tons of anthracite passed its length during the first year. Operations reached their peak in 1872 when 2.9 million tons were 2 This figure was 1,073 feet until the final improvement in 1875 when lock changes raised it. The difference in eleva- tion between terminals was 343 feet. moved.3 From that time, competition from an ex- panding railway network rendered the canal obsolete with increasing rapidity, tonnage gradually declining until final cessation and abandonment in 1898.4 Improvements and Enlargements When the canal was opened, it was the sole means for transporting coal out of the anthracite region. It 3 Noble Whitford in his History of the Canal System of the State of New York states that the peak year was 1868 with almost two million tons, but the higher figure and later year, from A Century of Progress—History of the Delaware & Hudson Company 1823-1923 are more likely correct. 4 The history of the Delaware & Hudson Canal has been well documented and related. The best account is Wakefield's extremely detailed, beautifully illustrated, and thoroughly enjoyable Coal Boats to Tidewater—the Story of the Dela- ware & Hudson Canal. See the bibliography for this and other works. NUMBER 10 was shallow—four feet in depth—with a waterline width of 28 feet (soon increased to 32 feet) and a bot- tom width of 20 feet. The first boats held 20 tons of coal. With a supply assured, the use of anthracite for heating, iron smelting, and steam generation ex- panded rapidly, engendering more business for the mines and canal. As a result of this cycle of prosperity, the canal eventually reached its capacity. Even with the introduction of 30-ton boats, by 1841 the demand for coal had so increased diat the canal's limit had been about reached. In that year, 192,000 tons were carried—27 times the first year's tonnage. The Delaware Aqueduct was built as an integral el- ement in an almost continuous program to increase the canal's capacity, therefore a brief survey of the various improvements will be useful for placing the aqueduct in its setting. The need for periodic enlarge- ments had been assumed almost from the outset, but as in the construction of other pioneer American transportation ventures like the Baltimore & Ohio Railroad, the modest capital initially available and the uncertainty of later needs dictated that the first route incorporate many expediencies and compromises. With the profits from the first decade's operation, it was possible to undertake a modest enlargement of the canal.5 In November 1842, at the close of the boating season, work was begun to deepen the trench to five feet by dredging the bottom and building up the bank height with the spoil, permitting passage of 40-ton boats. The happy fiscal effects of the project, com- pleted by 1844, were so pronounced that the canal management in 1845 began another increase—to pro- duce a 5/2-foot depth which would pass boats of 50- ton burden and result in an annual canal capacity of one half million tons. The cost of the project was $232,000. Even this enlargement was recognized as inade- quate practically before completion, for not only was the demand for coal increasing geometrically, but the progress of the Erie Railroad into the Delaware Val- ley and toward the coal regions in die mid 1840s an- 0 According to A Century of Progress the improvements were financed also by increased capitalization. Up to 1845 the share capital was $1.9 million. At the end of 1847— to finance the major enlargement then under way—it had been increased to $3.9 million, and by the end of 1850 to $6.6 million. The net profit for the year 1850 was about 12 percent on invested capital. nounced the end of the canal's monopoly of anthracite transportation. Consequently, the company was com- pelled to operate as economically as possible in order that its rates might be competitive with the railway's, if not actually lower. The only available means of re- ducing coal transportation costs between Honesdale and the Rondout depot were by increasing the capac- ity of the boats and reducing transit time. With the threat of competition from the Erie has- tening them into already inevitable action, the Dela- ware & Hudson directors in 1846 authorized the most ambitious enlargement project in the canal's history. The plan was to increase both capacity and speed, the former by both further deepening—to 6 feet—and widening, so that boats of 98 tons could be accommo- dated. The annual capacity would be thus drastically raised to one million tons, about five times the canal's 1842 capacity, an indication of the growing impor- tance of both anthracite and the canal in the coal in- dustry. The estimated cost was $1.1 million. The prin- cipal consequence of die widening was the necessity for rebuilding all locks and aqueducts, the former being enlarged from the original size of 9^2 feet by 75 feet to 15 by 90 feet. The lock-gate design was also changed to permit faster locking through. The most significant improvement to the canal's op- eration, however, was to be a material reduction in the passage time by removal of the worst bottleneck in the system—the slack-water crossing of the Delaware between Lackawaxen, Pennsylvania, and Minisink Ford, New York, just above the mouth of the Lacka- waxen. As capital originally had been inadequate to build an aqueduct for the purpose, a still pool had been formed by damming the Delaware, into which the boats were locked down on each bank. They then crossed the river either by momentum or hand haul- age along a ferry rope strung between the banks, the mules being carried over separately on a small rope ferry. Under ideal conditions the crossing was slow and a serious operational snag; at worst, during high water in spring and fall, the passage was impossible and canal operations came to a halt for days at a time. A further hazard was conflict with the consid- erable traffic of timber rafts on the river. The rafts- men, forced to traverse the low canal dam either by shooting it on the flowage over the crest or passing through a sluiceway, in general were understandably hostile to the canal interests and constantly engaged [Text continues on page 6] SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY DEVELOPMENT OF THE SUSPENSION AQUEDUCT DESIGN—PITTSBURGH The design employed by Roebling for all four Dela- ware & Hudson Canal aqueducts sprang forth fully developed in his first suspension structure, a 7-span aqueduct erected in 1844-1845 to carry the Pennsyl- vania State Canal over the Allegheny River at Pitts- burgh. The executed plan, however, evolved only after passing through a number of design stages. (Drawings courtesy Rensselaer Polytechnic Institute.) FIGURE 3.—As a means of achieving water- tightness, Roebling in all early schemes pro- posed to form the aqueduct trunk of wrought- iron plate, a rational choice in a city already a major iron center. By supporting the sus- pended structure at its extreme width, the floor beams acted as simply supported beams of con- siderable length—about 29 feet—with the great load of water bearing at their centers. The beams had thus to be of inordinate depth. Roebling obtained this by building up a 40-inch beam from a 16- and two 15-inch sticks, blocked apart by the longitudinal stringers. FIGURE 4.—Here the floor beams have been further stiffened by deepening to 46 inches and the addition of diagonal struts to transfer the trunk load more directly to the suspension points. t 1 U ' n n - • I u fry n : w t i pr rv *=t K\ ' // ' W // T \\ // in : M\ n : L- '€l\ L 1 //'\\ ■ // \\ 1 v/ . \\ ^E ..— - _!_:_.- -:,:- - : _i 1 m;NUMBER 10 FIGURE 5.—In this design the floor-beam members were sprung into a simpler double- bowstring form of 36-inch depth. While somewhat less stiff than the previous plans, the longitudinal spacing of the frames was almost halved, from 7 feet to 3 feet 10 inches, produc- ing greater overall strength. Roebling's predilection for the Egyptian Revival, ultimately manifested so strikingly in the magnificent stone towers of his Niagara Bridge (Figure 25), was first seen in several of the Pittsburgh preliminary designs. He estimated weights and costs for rendering in both marble and cast iron what he termed the "pyramids." »/,/., //. .',„//,„ FIGURE 6.—The design finally developed and accepted as " part of the agreement of 28 August 1844 ." bore but tenuous resemblance to its predecessors. The principal change and improvement was moving the trunk system's points of suspension in from the outer ends of the floor beams to points just outside the trunk sides, effectively reducing the bearing length of the beams from 28 feet to 18, and increasing their load-supporting capacity about two and a half times. Moreover, they then acted as con- tinuous beams. The weight of the towpaths and bracing, and the pressure of the water against the trunk sides acting through the inside diagonal struts, all bore downward on the cantilevered outer ends of the floor beams, materially counteracting the stress imposed by the water load at the center and further lowering the total stress in the beams. These transverse beams were finally reduced to pairs of 6 x 16s, spaced every four feet. The iron-plate trunk and the architecturally elaborated pyramids of iron or marble were casualties, presumably victims of harsh fiscal policy; but the double-diagonal wood-plank trunk sides and floor that replaced the iron added enormously to the vertical and lateral stiffness of the spans, and if cheaper, were certainly also better. All elements of the Pittsburgh Aqueduct were pro- portioned and disposed to perform economically as well as effectively, resulting in a design of high efficiency. In the Delaware & Hudson spans three years later, Roebling found it unnecessary to make any appreciable modification of the plan. The Pittsburgh Aqueduct served well until abandonment of the canal in 1860, following which it was removed. SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY // FIGURE 7.—R. F. Lord's plan for re-routing the canal at Lackawaxen in conjunction with an aqueduct crossing of the Delaware. Rough sketch sent to John A. Roebling 27 February 1847. (Courtesy of Rensselaer Polytechnic Institute.) [Text continued from page 3] the company in physical and legal harassment. An aq- ueduct had, in fact, been projected from the canal's beginning. The need now being pressing and the capi- tal available, it was included in the enlargement plan. Construction of the Delaware Aqueduct R. F. Lord, chief engineer of the canal, in planning the enlargement relocated the canal route at Lacka- waxen, establishing the aqueduct over the Delaware not above the mouth of the Lackawaxen River at the rope ferry site, but just below. This necessitated, in addition, construction of a second new aqueduct—over the Lackawaxen (Figures 7 and 8). Every Delaware & Hudson Canal scholar and author has speculated on Lord's reasons for planning the new route in that seemingly extravagant way, without having drawn any very convincing conclusions. There were obvious dis- advantages to the scheme, notably the added cost of the second aqueduct and the fact that the piers of the Delaware Aqueduct would be subject to the collective flow and battering of ice from both rivers. Two rea- sons are most commonly assumed for the re-routing: political considerations; and riverbed and riverbank conditions unfavorable to the upstream location. The first, in the case of a private company under the scru- tiny of its stockholders, seems unlikely, and there is nothing in the topography of the site lending much support to the second. More reasonable is a recent be- NUMBER 10 YORK PEN FIGURE 8.—The canal at Lackawaxen, about 1860, showing both the old canal and the new- route across the ''flats" between the new aqueducts. (Courtesy of Manville B. Wakefield, from Coal Boats to Tidewater.) lief of Manville B. Wakefield, author of the definitive Delaware & Hudson Canal history, that if the aque- duct had been built at the ferry, practically opposite the Lackawaxen's mouth, the piers would have been in constant jeopardy from the great ice floes that an- nually came down the Lackawaxen, grinding across the Delaware to the eastern shore with great force. Damage from these floes necessitated practically yearly repairs to the lock and bank of the canal which had been there before the aqueducts. Another likelihood, however, is suggested by the site conditions. Had the ferry location been selected, the aqueduct would have been right in the slack-water pool, with several consequences. First, there would have been less vertical clearance under the aqueduct for the rafts, probably an insufficient amount at spring high water when much of the rafting was done. Worse, the cofferdams used in building the aqueduct piers would have to have been considerably higher and heavier, and the entire problem of pier construc- tion would have been a good deal more difficult in the deeper water of the dammed pool, quite possibly to a degree more than offsetting the added cost of the Lackawaxen Aqueduct. There is also the probability that in the twenty years the Delaware had been stilled above the dam, quantities of silt had been deposited in the pool so that there would have been that much more material to excavate before reaching a solid footing. Finally, the river, in addition to being deeper. was, on the evidence of contemporary photographs, SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY T3 .S OJ —- P -° ZH <1J £ « 3 « Q > (L) J2 J3 « r>2 T3 C 3 fe O Pk C< o » iHHMM^»^»^4hM^tJHMHH)-»-*l|lSlrM^ l» K.l,».H.»..X-«-,»1«„t.|t,jt..H,«.»,-)r,t-jt,«.».,IHHH|7f IN 1698, THE AQUEDUCT WAS DEWATERED AND CONVERTED INTO A HIGHWAY TOLL BRIDGE WHICH FUNCTION IT CONTINUES TO SERVE. THE WOOD TRUNK WAS REPLACED BY THE PRESENT DECK SYSTEM FOLLOWING A FIRE IN 1932. SCALE M faer A/. W. ELEVATION- SECT/OK MUWN wr LHJC i^cOVr . SJiEE 7 I or 3- 7/V/5 STffocruee MOHAWK-HUDSON AREA SURVEY DELAWARE AND HUDSON CANAL- DELAWARE AQUEDUCT NY 5529 MINISINK FORD, SULLIVAN COUNTY. NEW YORK LACKAWAXEN, PIKE COUNTY, PENNSYLVANIA I HISTORIC AMERICAN BUILDINGS SURVEY HT 16 OF 20 aun FIGURE 45. FIGURES 45, 46, and 47 (overleaf).—The following three drawings were made in August 1969 by the Mohawk- Hudson Area Survey (M-HAS), sponsored by the Ameri- can Society of Civil Engineers, the National Park Service, the New York State Historic Trust, and the Smithsonian Institution. The M-HAS was the first project of the re- cently formed Historic American Engineering Record, established to prepare and preserve graphic records of significant American engineering monuments. NUMBER 1 0 35 FIGURE 46. 36 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY NUMBER 10 37 FIGURE 48.—The New York shore from Lackawaxen. THE DELAWARE AQUEDUCT TODAY, Figures 48-52. (Photographs by David Plowden.) FIGURES 49 and 50.—Traffic during most of the year is steady though light, but dwindles from January to March. As the only crossing of the Delaware for fifteen miles, the bridge fills a decided local need as well as being unique point of interest. 38 SMITHSONIAN STUDIES IN HISTORY AND TECHNOLOGY FIGURES 51 and 52.—The contrasting massiveness of masonry piers and lightness of wire cables, so much the measure of the suspension bridge and so often extolled by poet and painter in the Brooklyn Bridge, is as fully marked at the aqueduct. FIGURES 53.—Essex-Merrimack Bridge near Newburyport, Massachusetts, before and after. In the 1909 "rebuilding" of the 1810 structure, the entire superstructure was replaced with a loose replica, leaving of the original fabric only the pier masonry below deck level. (From Engineering News (25 September 1913), volume 70, page 585.) FIGURE 54.—The "Wire Bridge," New Portland, Maine. While having undergone some rebuilding, the bridge is original in its principal elements and is a rare survival of an early suspension structure. (Photograph by David Plowden.) LIMBER WJHCTED, on THE DEL. & HUDSOJV CAKAL. Will be received until the 10th day of March next, for furnishing and delivering the following bill of Lumber, viz : St fe*t long, 1 by 7 inches at one end, 7 by 7 at the other,—16 ft. long, Ah 7 by IS,—20 feet long, do, 2 1-3 by 10,—10 feet long, do. 2 by 10,—7 feet 8 indies long, do. 7 by 7,—12 feet long, do. 6 by 7,—6 feet 8 inches long, 1,600 feet Linial Measure, 7 by 7 inches, for Bailing, i'l.i.nu 8,600 6,533 1,600 feet Linial Meuore, 6 by 6 inches, for Bailing, 1,400 do. 6 by 7, any length orer 20 feet, 1,400 do. 5 by 5. do. 1,450 I'lunk, 25 or 26 feet long, 2 1-2 inches uniformly thick, 76,680 Flank, 14 feet 4 inches long, 2 1-2 inches uniformly thick, 76,680 . by 10, or 2 inches by 12, either 16, 20, or 24 ft long, 22,400 2 by 10 inches, 16 or 24 feet long, 19,200 Jo of Pii Total Board Mci All the above bill to be of good sound White Pine, and work full size, free of shakes, rents or blaek knots, when counter-hewed, and delivered on the Pennsylvania side of the Delaware river above high water mark, between the mouth of the Lackawaxen and Delaware Dam (for Del, and Uud- Canal) by or lief in" the first dij of Jul v next. Payment will be made when the Lumber is de- livered on the bank as above stated, and approved and accepted lo (he satisfaction of the Knginecr on Delaware and Hudson Canal for the time being. Proposals are desired to be i » writing, stating the price per one thousand feet board measure, and directed to the subscriber, at the office of the Delaware and Hudson Canal Company, in llonesdale, Wayne county, Pa. For any information relating to the above bill of Lumber, apply to t