Manassas Gap Railroad Engineering
by Debbie Robison
December 2002
By the time the Manassas Gap Railroad was built in the 1850s, engineering concepts had been developed for the construction of railroads. Mathematical theory was applied to most aspects of roadbed, embankment, and culvert design. Information contained in this article regarding actual construction of the Manassas Gap Railroad pertains to culverts and embankments in Annandale, Virginia.


Engineering calculations to determine the steepest railroad roadbed grade were based on economics, rather than mere practicality. Engineers felt that a train could pull fewer tons of goods up a slope towards market. It was originally believed that heavier grades had a proportionately greater impact on the weight a freight train could pull; however, by 1871, engineers determined that heavier grades were less injurious since so much of a greater ascent was required to double resistance. (Resistance was a primary factor in the equation used to calculate required locomotive power.)

In consequence of the engineering theory of the day; however, Manassas Gap Railroad engineers designed roadbeds with minimal slope by choosing a flat route, when possible, and both trenching though hillsides and building up embankments. Construction costs were greater on the sections of railway that required excavations and embankments. Historically, engineers attempted to minimize costs by utilizing excess excavated terrain, called “surplus”, in nearby embankments. If an embankment was not being built within a short distance, the surplus was deposited on nearby land in mounds called “spoil-banks”. When an embankment was constructed and there was no surplus available, the deficiency was called “wantage”. The required terrain was then obtained from “side-cuttings”. (Several probable side-cutting sites were located within the Manassas Gap Railroad Historic Site.) The excavation of earth was facilitated by the use of picks, ploughs, scrapers, and spades to loosen the earth prior to shoveling the dirt and rocks into wheelbarrows and horse-drawn carts. Steam-powered machines were also used at this time for excavation and could dig and load 1,000 cubic yards of earth per day. Where wheelbarrows were utilized, planks were laid in a path from the side-cutting or excavation area to an off-loading area. It was recommended that the slope leaving a side-cutting site not exceed 1 in 12.

Generally speaking, once a route was established, the boundaries of the roadbed were staked. Specifications may have been provided requiring the area be cleared of “all trees, logs, brush, and other vegetable matter.” All stumps and roots should then have been grubbed out, followed by the removal of two feet of topsoil.

Precautions were necessary in constructing embankments to ensure they would not slip due to both their size and the vibrations engendered by passing trains. The ideal method, albeit more costly in time and expense, of building embankments was to form the embankment in layers, or courses, not to exceed three or four feet thick. (This method was especially recommended around masonry.) Vehicles used to convey the material were required to drive over each successive layer to compress the soil. Sometimes engineers found it advantageous to have the soil rammed. Ideally, the layers were deposited in a concave fashion, thus reducing the likelihood the soil would slip down the side slope. One practice used for reducing this slip was to initially form the embankment wider at the top and narrower at the bottom. The outside edges of the embankment were formed first; then the center was gradually filled so that the earth had a tendency to move towards the middle of the embankment. Finally, the excess earth at the top of the embankment, as a result of forming the embankment wider at the top, was thrown down the slope to fill in the narrow bottom portion of the embankment. During construction, a trench may have been built at the foot, or “toe”, of an embankment to reduce the tendency of the embankment to spread. Engineers designed the side slope of embankments based upon the type of soil utilized.  It was believed that common earth would stand on a 1 to 1 (45 degree) slope, though a 1-1/2 to 1 (33 degree) slope was preferred, especially for high embankments. (A 33 degree slope was achieved for the Manassas Gap Railroad Historic Site embankment, which is composed of silty, slightly micaceous fine sand and rock.) Gravel and high-clay soils required shallower slopes. Slopes for excavations were recommended at 2 to 1 to allow sun and wind to reach the roadbed to keep it dry.

The predominant embankment construction method involved building up the embankment to its full height at one end and continuing the roadbed by transporting wagonloads of earth (containing about three cubic yards) atop the embankment along temporary rails to its current terminus where it was dumped. It was a recommended practice to seed the side-slopes of both embankments and excavations to stabilize those areas.

The top of embankments and the bottom of excavations were brought to a height two feet below the intended finished height, or "formation level", though allowances were made for settling, or “shrinkage”, of embankments. The surface was shaped with a crown at the middle with a falloff to the sides, similar to road construction, to allow water runoff. Ballast, composed of gravel, broken stones, or quarry rubbish, was set atop the formation level to spread the bearing of the sleepers over a large surface of ground, keep the track in place, secure drainage, and give medium elasticity. The term ballast was derived from the original use of ships' ballast. The ballast was laid on rock as well as earth and, if properly placed, extended the full width of the top of the embankment. Recent inspection of the embankment that passes through the Manassas Gap Railroad Historic Site does not reveal the presence of ballast. Either the embankment only reached the formation level when construction of the Independent Line ceased in 1857, or the Manassas Gap Railroad Company intended to lay the rail directly on plain roadbed. Company records indicate that rail was laid on both ballast and plain roadbed.

The width of the roadbed included the distance between the rails and an area on each side of the rails, called “side-spaces”, used in the event of train derailment. The side-space widths were from 5 to 8 feet for different railways. High embankments were given wider side-spaces. The design width of a roadbed was determined by adding the width between the rails to the side-space widths.  The gauge on Virginia railways was five feet; therefore, roadbeds were probably designed from 15 to 21 feet wide, for single-track routes. Based on the design criteria, the roadbed passing through the Manassas Gap Railroad Historic Site was engineered for a single track.


Manassas Gap Western Culvert OutputWhile Manassas Gap Railroad Chief Engineer John McD. Goldsborough may have had confidence in his engineering theory and practice, today we know with certainty that the culverts built by the Manassas Gap Railroad Company have proven Goldsborough's ability by standing the test of time.

The culvert foundations were made of large 18" to 22" quarried micaceous gneiss blocks. Engineers were aware in the mid-nineteenth century that ineffective culvert foundations caused the superstructure to sink into the soil below, especially under the middle of the embankment. The foundation and flooring stones were sloped to facilitate water drainage. A slope of 1” for every 10’ was desired. At the Manassas Gap Railroad Historic Site, a slope of approximately 7/8” in 10’ was achieved at the western culvert and approximately 1-1/4” in 10’ at the eastern culvert.

The culvert walls were made of stones, approximately 36" to 40" thick, stacked on top of the foundation stones in a straight line and plumb. As the culvert wall face stones were stacked, large stones were set behind them. Smaller stones were used as fill to level the back stones with the face stones. Once both walls of the culvert were constructed to finished height, very large lintel stones were set over the walls to span the culvert. These stones bear on both walls a minimum of 6" and are from 12" to 20" thick. The lintels were most likely hewn by sledge hammering a star bit to an average depth of 2-1/4” and boring additional holes along a line spaced, on average, 6” apart. An alternative tool for drilling holes was a long steel bar, chisel-edged, which was raised and let fall on the desired point. At each stroke, the bar was partially turned so the cuts crossed in a star pattern. Some masons inserted feathers, half-round metal pieces, into the holes with wedges set in between. The wedges received even pressure along the line to split the rock. Other masons dripped water onto wooden dowels placed in the holes. Expansion cracked the stone. Explosives were also employed to blow apart the rock. This “shot” stone typically contains hairline cracks and is not ideal for masonry.

Manassas Gap RR lintelsThe lintels were shaped into an elongated pentagon configuration with a triangular, roof-like peak. This resulted in the bearing pressure of the embankment to be distributed over the lintel, the back stones, and the fill stones in the same manner that an arch bears weight. Evidence exists at the western inlet opening that small stones were stacked in an arch-shaped manner over the culvert. This design, along with the firm foundation, kept the embankment from pressing the side walls together, a concern that nineteenth century engineers designed to overcome either by using a large masonry foundation or an inverted brick arch flooring.

Historically, when soil conditions were marshy or of "quicksand", foundations were laid on a 3' to 6' thick bed of gravel, sand, or "stone broken to turnpike size." An engineering text recommended the bed extend several feet beyond the masonry in each direction and rammed to compact the fill. The gravel and sand were thoroughly wet to aid in the consolidation. It was believed this foundation bed was sufficient for a "moderate height embankment."

When masonry was laid on a smooth surface of rock or a platform of timbers, stone pilings were driven in front of the masonry to prevent slippage. No evidence of pilings was discovered at either of the Manassas Gap Railroad Historic Site culverts; however, an apron, composed of stone rip rap, was placed adjacent to the masonry to prevent undermining by scour. Sharp directional changes of Coon Branch at the culvert outlets most likely caused foundation undermining as high water exiting the culvert struck the opposite bank and caused the water to move in a swirling motion back to the foundation.

Each end of the culverts was secured with large headwalls. These walls have finished faces, not polished, with joints from 1/2" to 1" thick. The stones have cut corners, level tops, and like the rest of the culvert masonry, were laid completely dry. Portland cement was not available in the mid-Atlantic region until 1872. The original builders would have concluded that a lime-based mortar could not have withstood the effects of constant water erosion. Careful examination of the masonry shows no evidence of mortar. Built properly, stone walls have the ability to move water away from themselves and therefore do not suffer the effects of a freeze/thaw cycle.

The stone culverts and earthen embankments at the Manassas Gap Railroad Historic Site were constructed using generally accepted civil engineering practices of the mid-nineteenth century, many of which continue to be used today.


Cochran, Allen. Telephone interview. 5 November 2002.


Gilespie, William. M., LL.D., C.E. A Manual of the Principles and Practice of Road-Making: Comprising the Location, Construction, and Improvement of Roads: (Common, McAdam, Paved, Plank, Etc.) and Rail-Roads., Tenth edition, With Large Addenda. New York: A.S. Barnes & Company, 1871. Virginia Historical Society.


McD Goldsborough, John. Letter to the Secretary of the Virginia Department of Public Works. Statement of Prices Paid for Sundry Kinds of Work. 22 March 1859. Library of Virginia, Manassas Gap Railroad Collection.


McRaven Charles. Stonework techniques and Projects. Massachusetts: Storey Books, 1997, p. 55.


Trautwine, John C., The Civil Engineer’s Pocket-Book of Mensuration, Trigonometry, Surveying, Hydraulics, Etc. Phil, Claxton, Remsen, & Haffelfinger, 1874. Virginia Historical Society.