by Robert McCullough

    The transportation agency will be developing an interpretive display for the Williamsville Covered Bridge as part of the reconstruction. Consequently, I've been digging into town records trying to gather as much information as possible. The date of construction and builder remain a mystery, but I've suggested a likely explanation. I'm attaching the material just to pass along the information, for what it's worth. I really need another trip to Newfane to dig into the handwritten records prior to 1850, but the town clerk's office is in transition and records are in disarray.

    Williamsville is situated along the first branch of the West River, now the Rock River, but for many decades called simply "the Branch." Both the McClellan Map of Windham County (1856) and the Beer's Atlas of Windham County, (1869), identify three bridges in Williamsville. One crosses the Branch at the village's easterly end, where a covered bridge stood until destroyed by fire and replaced by a reinforced-concrete arch bridge in 1908. The second bridge crosses Baker Brook westerly of the village center on the road to Dover. The Williamsville Covered Bridge is the third crossing, also located on the road to Dover a short distance westerly of the bridge across Baker Brook. Although the builder and precise date of construction for the Williamsville Covered Bridge are not known, it was probably erected no later than 1860.

    Newfane's handwritten records of town meetings and printed annual reports, the latter beginning in 1860, are inconclusive regarding the date of construction, but various clues point to the decade of the 1850s. By then, the improvement of roads leading to and from Williamsville had become a frequent topic at town meetings, possibly the result of periodic flooding. The 1856 McClellan map shows a mill pond at the site of the Williamsville Covered Bridge and also an adjacent pail factory. However, that pond and factory do not appear on the Beer's atlas in 1869. Instead, the Beer's map shows a mill pond crossed by the bridge at Baker Brook and, at the pond's outlet, a grist mill. A flood of 1856 did wash away several buildings and bridges along the Branch, but whether flooding caused the loss of the earlier pond, dam, and factory, or whether one of the maps is incorrect, is not known. Another freshet in July, 1859 destroyed nearly every bridge on Baker Brook,, as well as a saw mill owned by Dana D. Dickinson and E. P. Wheeler on that same stream.

    Written accounts of town meetings up to 1859 do not mention the Williamsville Covered Bridge by that name. In addition, no printed copies of the First Annual Report (1860) have been found. However, the Second Annual Report (1861) discloses that L. Halladay received $5.93 for painting the Williamsville Bridge. Five years later, sometime during 1865 or early 1866, E. P. Wheeler received $5.00 for unspecified labor on a bridge by that same name. In 1877, the town spent $113.03 to repair the .Williamsville Bridge. and to apply new shingles. Twenty years later, 1897 or early 1898, the town paid L. O. Morse $106.00 to install metal roofing, again using the name Williamsville Bridge. Unfortunately, whether these activities refer to the Williamsville Covered Bridge or to the covered bridge on the village's easterly end cannot be verified. Identification of the builder is also not possible, but timber framers erected a number of bridges in Williamsville and nearby Pondville (South Newfane) during these years of road-building, and the names Daniel O. Stratton, E. P. Wheeler, Dana D. Dickinson, B. E. Morse, and C. W. Morse appear frequently in town records.

Structural Deficiency
    The Williamsville Covered Bridge is located on a busy Class II road used by heavy emergency vehicles and school buses, and no convenient alternative routes are located nearby. As originally designed, the bridge's 118-foot Town lattice truss proved to be inadequate for the weight of modern emergency vehicles. Many of its timber components were comparatively small when measured against other bridges of this design and span length. Moreover, many modifications had been introduced over the years to strengthen the structure, including numerous sister lattice planks. At one point, the original timber floor beams had been replaced with steel I-beams, which were eventually removed in 1979. That year, the lower five feet of all lattice planks were replaced, but with chords spliced by bolts. Unfortunately, the modified lattice components lacked sufficient rigidity, creating ere severed from each truss lattice and replaced with seven-foot segments spliced to the remaining lattice with bolts. Upper and lower bottom chords were also a systemic weakness in the trusses. By 1998, the truss lattice and top chords revealed wracking, bowing, and negative camber, indicating that the entire structure had begun to fail. As a temporary means for keeping the bridge open to traffic, engineers installed two very large glulaminated girders inside the structure that year, one on each side of the single travel lane.

Reconstruction
    In 2001, the Vermont Historic Covered Bridge Committee began discussions about the best method for preserving the Williamsville Covered Bridge, observing the guidelines specified by the Vermont Historic Covered Bridge Preservation Plan. That plan seeks two primary objectives: (1) to preserve the historic physical and structural integrity of covered bridges to the maximum extent possible; and (2) to retain covered bridges for use on the state's network of roads whenever possible, thus preserving the historic functional integrity of these bridges. The plan establishes priorities for ten preservation treatments, and the committee must establish that a preferred treatment is not possible before moving to the next recommended treatment.

    The Williamsville Covered Bridge is a case study showing that achieving these worthy objectives can be very difficult. In its completed form, the bridge is an accurate reconstruction of the original 19th century design, with only nominal changes to the overall dimensions and to the size of some chord members. Those changes include glu-laminated lower bottom chords to increase the bridge's strength, one of several methods of strengthening considered by the committee.

    The preservation plan recommends traditional methods of strengthening, and the committee initially approved a plan to restore the bridge to its original design, replacing deteriorated components in kind, and then adding a new, lattice truss to the outside of each existing truss. The committee recognized the experimental nature of this proposal but justified its potential compromise to the bridge's structural and visual integrity, reasoning that engineers would be able to test the way the structure functioned and thus evaluate its applicability at other sites. However, committee members eventually rejected this plan due principally to the resulting visual compromises to the bridge and because so many of the bridge's existing components would require replacement in any event.

    The plan also permits the introduction of large glu-laminated girders as a co-functional, reversible supplemental structural system, reinforcing the existing structure. Two conditions must be satisfied to adopt this treatment: (a) other preferable treatments must be unworkable; and (b) the bridge must be restored to its original form. The committee investigated the possibility of installing two large glu-laminated girders beneath the existing structure, thereby satisfying loading requirements. However, the substantial depth of those beams created poor clearance above the river.

    Various other treatments permit the committee to modify, to a very limited extent, the size of the bridge's individual structural components in order to achieve the desired load capacity, specifying that changes to the bridge's overall dimensions should not occur. Recognizing that almost all of the bridge's structural components would require replacement, the committee elected to reconstruct the entire bridge and to modify the original design only to the extent required to carry emergency vehicles.

    The reconstructed bridge standing today differs only slightly from what engineers and historians believe was the original truss design. The overall bridge width increased by a total of approximately eight inches and bridge height (from upper top chords to lower bottom chords) decreased by two-and-one-half inches. The paired sets of planks, four each for upper and lower top chords and for upper bottom chords, were increased in depth by one-half inch. The sets of four planks for the lower bottom chords were replaced by sets of two glu-laminated beams, six-and-three-quarters inches by twelve-and-three-eighths inches, one on each side of the truss web. The dimensions of the original floor beams are not known, but those installed in 1979, to replace steel I-beams, measured twelve-inches-by-eighteen inches set at four-foot intervals, center to center, and may have been larger than the original floor beams. However, many of the 1979 floor beams were replaced with glulaminated beams in 1998 when the temporary glulaminated girders were installed. The glu-laminated floor beams used in the reconstruction, ten-and-one-half-inches by fifteen-and-one-eighth-inches set at four-foot intervals, are slightly smaller than the floor beams used in 1979. The lattice planks of the trusses are identical in dimension to those used in the original design, fifteen-feet by ten-and-three-quarter-inches by three-inches.

[Robert McCullough, of the UVM Dept. of History and member of the Vtrans Historic Bridge Committee, is Author of Crossings, A History of Vermont Bridges VAOT & VHS, 2005 - Ed.]
[* reprinted from THE BRIDGER, Summer 2009]

Return to top