Fortifications Images

Manassas Junction, Virginia. Confederate Fortifications, East Front, Interior Line.

Manassas Junction, Virginia. Confederate Fortifications, East Front, Interior Line of Open Detached Field Works.

As Shown In:

Barnard, George N. Manassas, Va. Confederate Fortifications, with Federal Soldiers. March, 1862. Brady Civil War Photograph Collection (Library of Congress). http://hdl.loc.gov/loc.pnp/cwpb.03910 (Last Access Check: 04/08/02). Noted as 03910u.

Barnard, George N., Manassas, Virginia. Fortifications. March,1862. Brady Civil War Photograph Collection (Library of Congress). http://hdl.loc.gov/loc.pnp/cwpb.00102 (Last Access Check: 07/08/04). Noted as 00102u.

Barnard, George N., Manassas, Va. Confederate Fortifications, with Federal Soldiers. March, 1862. Brady Civil War Photograph Collection (Library of Congress). http://hdl.loc.gov/loc.pnp/cwpb.04332 (Last Access Check: 07/08/04). Noted as 04332u.

O'Sullivan, Timothy H., Manassas, Virginia. Eastern Range of Confederate Defenses. July, 1862. Brady Civil War Photograph Collection (Library of Congress). http://hdl.loc.gov/loc.pnp/cwpb.01084 (Last Access Check: 07/08/04). Noted as 01084u.

[Please Note: Images presented here are for illustration purposes only. These are not the original images of photographs discussed in the text and are not the images that these observations are based on. Images are referenced by the Library of Congress filenames of the original images of the photographs downloaded from the Library of Congress web site.]

Original Image Courtesy Library of Congress

This photograph shows four field works constructed by Confederate forces to defend Manassas Junction, Virginia according to a plan developed by engineer Thomas H. Williamson in the spring of 1861. Williamson's plan for the defense (which may have been later modified by P.G.T. Beauregard and his staff) of the vital rail junction and supply depot envisioned an interior ring of open detached field works that covered all avenues of Original Map Image Courtesy United States Coast Surveyapproach that an enemy might take to reach the junction. An advanced line of enclosed detached field works guarded the distant approaches to the supply depot from the direction of Centreville and Bull Run Creek. These four works are shown in the detail from the Whiting Map of Manassas Junction to the right. They extended from just south of the Orange and Alexandria Railroad in a roughly northwest direction for a distance of about 360 yards.

Several features can be viewed in sufficient detail in this (and other photographs which will be introduced as necessary) to allow some effective comment on the construction of these field works. First, the interior details of the battery south of the railroad from which this photograph was taken (marked A on the map); second the profile and exterior details of the first redan on the north side of the railroad along with its odd barrel traverses (marked as B on the map); and third, both interior revetment and exterior details of the large redan in the distant background that covered the salient angle joining the east and north fronts of the interior line of works (marked D on the map). The smaller battery (marked C on the map) between the first and second works north of the railroad can not be seen with sufficient clarity to make any but the most cursory comments on its trace or structural substance.

I. Battery Interior (A)

A. Original Image Courtesy Library of CongressThis battery was positioned on the left face and salient of an irregular redan immediately south of the Orange and Alexandria Railroad where its embrasured naval ordnance could fire both across the front of collateral works to the north and eastward along the line of the railroad tracks. Both the basic materials used to construct this battery and the arrangement of the revetment of the interior slope can be seen in the foreground. This is a split revetment that combined the post and timber revetment method for sustaining the lower section of the interior slope with a rough hurdle work to sustain the upper section of the slope. Posts were fixed in the ground at somewhat irregular intervals averaging about 3 feet from post to post along the foot of the interior slope. The lower section of the revetment was constructed A. Original Image Courtesy Library of Congressby laying irregular split and cut timber between these posts and packing the earth of the parapet against them to bond the revetment. When this revetment reached a height of about 2 and one-half feet rough hurdle work consisting of the larger unstripped branches of cedar trees were run in and out between the posts. As the soil was thrown onto the parapet small leafy branches were laid on the rising parapet to both decrease the amount of soil required to construct the parapet and better bond the somewhat rocky soil into a solid mass. This also tended to increase the parapet's resistance to the erosive effects of wind and rain.

A. Original Image Courtesy Library of CongressMuch of the revetment shown in this photograph (which was made about a year after these works had been completed) had suffered quite a bit of deterioration and was in a generally dilapidated condition. To the right of the embrasure most of the small branches holding the interior slope in place have rotted away or suffered other debilitating damage and the soil is beginning to fall and form a mound at the foot of the interior slope. Still, most of the revetment posts seem to be well fixed and many of the larger branches of the hurdle work A. 00102u. Original Image Courtesy Library of Congresscontinue to sustain the soil. Another photograph (00102u) shows that the section of the revetment immediately to the right of the revetment shown in this photograph had completely rotted away and the interior slope had fallen onto the foot of the interior slope and gun platform. To the left of the embrasure the small branches of the hurdle work were giving way when the photograph was made; weight of the parapet's rocky soil was bending the surviving branches and beginning to rotate one of the posts backward and to one side.

Most of the batteries in the Manassas defenses were designed specifically for handling heavy naval ordnance mounted on ship's trucks rather than standard land service carriages. In this case the embrasure had a A. 03910u. Original Image Courtesy Library of Congressgenouilliere about 2 and one-half feet above the foot of the interior slope that was sustained by a vertical post and timber revetment. A short vertical post at the center of the embrasure genouilliere under the embrasure neck marks an angle in the work where the line of the interior crest changes direction. This would have allowed a gun mounted at this embrasure a slightly wider field of fire than normal, but also required that a short horizontal huerter be placed on the terre-plein to prevent the gun truck from being run to far into the embrasure. The four horizontal posts, two on either side of the embrasure, that are not part of the revetment were tackle posts. Dark circles in these posts mark the holes where eye-bolts were once attached. Breeching tackle used to maneuver a gun mounted on a ship's truck and catch the gun as it recoiled when fired was run through the eye-bolts on these tackle posts and attached to the truck. This confirms reports that naval ordnance captured at the Norfolk Navy Yard actually was mounted in the Manassas defenses.

Whatever revetment may have originally been used to sustain the vertical slope of the embrasure cheeks had A. 03910u. Original Image Courtesy Library of Congresslong since disappeared by the time this photograph was made. Here the earth of the superior slope of the parapet can be seen beginning to slide down into the embrasure (underneath the man sitting on the parapet with his legs stretched out toward the embrasure), but the cheeks show very little sign of deterioration of their vertical angle. This was due to the bonded composition of the parapet; small branches that look like a root system can be seen sticking out from the cheeks. This of mixture of leafy branches and earth tended to hold the soil of the parapet in place longer and prevented it from washing out too quickly.

A. 03910u. Original Image Courtesy Library of CongressPlanks of the gun platform can be seen below and on either side of the embrasure. This consisted of 17 planks of varying width laid parallel to the directrix of the embrasure. Although the method used to give these deck planks a solid foundation can not be seen, whatever method was used seems to have worked exceptionally well since the planks seem to have retained a A. 00102u. Original Image Courtesy Library of Congresslevel surface without any edge plank displacement or warping and bowing of the individual planks. Another photograph (00102u) shows that these planks were not all the same length; rather, planks were cut to fit nicely flush with the decking of other gun platforms to create a smooth surface multi-gun platform on the terre-plein of the battery.

II. Redan with Battery at the Salient (B)

Infantry Parapet of the Right Face

B. 03910u. Original Image Courtesy Library of CongressNext in line on the north side of the Orange and Alexandria Railroad was a redan that combined parapets designed for a musket armament on the faces with a curved embrasured battery for three guns at the salient. Structural features of this work's infantry profile can be fairly well described since the end of the parapet is clearly visible next to the railroad tracks. This is a breastwork parapet without a raised banquette or banquette slope, but with other features common to this type of parapet well defined. The line of the interior crest can be traced from the end of the infantry parapet to the raised parapet of the salient battery by following the succession of revetment posts. The nature of the revetment is suggested by the horizontal timbers held in place by slightly sloped posts that are visible at the extremity of the parapet next the tracks. The superior slope seems to be about 3 feet thick at the top and rather flat without a distinct plongee. B. 03910u. Original Image Courtesy Library of CongressThe line of the exterior crest is very distinct and shows that the parapet had a very equal thickness throughout its length. The exterior slope falls at what appears to be about a 45 degree angle toward the ditch without intervention of a berm. Lines and slopes of the ditch are rather indistinct; width at the top was about 12 to 16 feet, but the maximum depth was probably only 3 to 4 and one-half feet. Still, the height of the parapet combined with the ditch would have made this a significant obstacle for assaulting troops to climb over under fire. One section of the exterior slope seems to be sliding into the ditch while another section, immediately adjoining the raised parapet of the battery, shows the B. 03910u. Original Image Courtesy Library of Congressdistinct line of the natural level of the ground. One unusual feature exhibited by the profile of this parapet is the presence of a slight epaulment immediately in rear of the banquette that runs nearly the full length of the parapet to the salient battery. This feature seems a bit too low to provide effective protection from bursting shells or reverse fire; it would certainly be an obstacle to reserves advancing to support troops manning the parapet. One has to wonder whether this might have been intended as a psychological barrier intended to help keep untested troops under fire for the first time at their posts on the parapet.

Salient Battery

B. 03910u. Original Image Courtesy Library of CongressThe distinct lines of the infantry parapet terminate rather abruptly as they intersect the lines of raised parapet of the salient battery. The interior crest of the battery parapet was about 2/3 higher than the adjoining infantry parapet and at least twice as thick at the top; the exterior slope falls at about a 45 degree angle onto a narrow berm. Although the ditch in front of the battery parapet can be seen clearly, it seems to have combined a rather sharply angled scarp with a gently sloped counterscarp. It would probably have been fairly easy for an assaulting body of troops to enter the ditch in front of the salient battery, but it would have been quite difficult for attacking troops to climb the scarp and scale the parapet.

As to the structural details of this battery, it appears to be constructed in much the same fashion as the battery south of the railroad. The tops of a series of posts can be seen rising above the interior crest and some branches can be seen jutting out from the slope connecting to the infantry parapet. Another photograph (03442u) taken from the interior of this battery shows that it too combined a low post and timber revetment with hurdle work and was arranged with tackle posts for maneuvering naval ordnance on ship's trucks. ThatB. 04332u. Original Image Courtesy Library of Congress photograph also shows the ends of branches sticking out from the revetment that have been buried in the body of parapet to bind the rocky soil into a single mass.

Photograph 03442 shows some of the battery's interior details that can not seen in the photograph under current consideration. This photograph confirms that the raised battery parapet had a split revetment that was in much better condition when the photograph was made than that of the battery south of the railroad. Many small cut branches can be sticking out of the rough hurdlework revetment, again confirming that the earth parapet was stabilized by inclusion of leafy branches buried in the body of the parapet near the interior slope. Again, tackle posts are present showing that this battery was also designed for an armament of heavy naval ordnance mounted on ship's trucks. A heurter was noticed at the foot of the interior slope of the battery south of the railroad; the embrasure shown in this photograph had a stopping beam suspended between two short posts inside the tackle posts nearest the mouth of the embrasure. This beam would catch a truck on its forward cheeks and prevent it from running into the interior slope.

Some construction details of the gun platform are visible in this photograph (03442u). The platform is rectangular with 15 uniformly sized deck planks laid with their length perpendicular to the direction of the B. 04332u. Original Image Courtesy Library of Congressembrasure mouth. Planks were laid on a foundation consisting of hewn timber sleepers covered with a crossing layer of thin planks. This platform was sloped upward toward the rear, a common feature designed to assist in shortening the recoil distance of the piece when discharged and to promote good drainage. A small rectangular addition to the platform can be seen in the lower right foreground of the photograph. This addition seems to have two planks affixed to it that run perpendicular to the direction of the interior slope of the parapet; these may have been a rather ad hoc stand for a pile of ready ammunition.

B. 04332u. Original Image Courtesy Library of CongressGround immediately in rear of the platform slopes downward at the natural of the soil and falls into a small drainage trench that runs along the rear of the salient battery. This seems to indicate that this was probably a cavalier battery raised above the natural level of the site. It may be assumed that this particular arrangement was designed specifically for the purpose of providing the structural means to give naval ordnance its full effect in a foreign environment.

Cask Paradoses

B. 03910u. Original Image Courtesy Library of CongressTwo of the more interesting fortification features visible in the photograph (03910u) are the two cask traverses immediately to the left (rear) of the salient battery. Although these appear at first glance to be closely associated with the battery, they were not integral elements of the battery, but were somewhat behind and to either side of the raised terre-plein of the battery. Instead, these were bullet-proof defilade traverses that extended across the terre-parade of the redan and, as such, can be classified as parados intended to screen the interior of the work and the rear of the redan faces from the two points of commanding ground on either flank of the redan; these two points being the battery south of the railroad and the most northerly redan farthest from the camera. The second traverse B. 03910u. Original Image Courtesy Library of Congress distant from the camera does, in fact, screen the interior slope of the opposite face of the redan from the camera's view. Except, of course, for those points where the casks have been displaced or were not positioned properly in the first place which allowed the camera to see the post and timber revetment of the work's interior slope.

Structurally, these traverses were laid out in much the same fashion as gabion traverses with the exception that the lower tier of the traverses was only two casks wide and the upper tier only one cask wide. It seems doubtful whether these traverses had the mass necessary to withstand any fire larger than that of a common B. 04332u. Original Image Courtesy Library of Congressmusket. Certainly the method of construction did not promote stability of the traverse. The casks of the lower tier have been laid side to side in two rows; in the far traverse earth was excavated from a wide, but shallow, trench and used to fill each cask and heap the soil on top of them to form a bed for the upper tier which was also filled with earth. Construction of the near traverse did not reach this point: the casks have been filled, but large gaps between the casks are visible, indicating that the voids between the casks were not filled. This would have been positively necessary to convert the bottom row from a series of independent elements into a unified mass.

Upper tier casks seem to have been laid either along the center line of the two rows of the lower tier or directly on top casks of the lower tier, but, judging from the number of upper tier casks that have been displaced or knocked completely off the traverses, they were not anchored into place. Upper tier casks revet and bond the traverse in the same way that a glass holds water, that is, not at all. Solid shot striking a cask of the upper tier would probably shatter the cask staves (sending wood splinters flying in all directions) and knock the cask completely out of place. In short, these interesting features reveal the hope and inexperience of their designer, but would not have been particularly useful for the defense of this field work. Once one of the casks had been shot away, it seems probable that troops manning the parapet in front of it would lose a bit of their confidence in the protective value of this field work.

Turning again to the closer view of this redan provided by the second photograph (03442u) some sections of the revetment of the work's left (north) face parapet can be seen through breaks in the cask traverse. This line of parapet had a post and split timber revetment. Posts are set at an appropriate angle to give the interior its necessary slope, but the split timbers give the revetment a rather discordant and ad hoc appearance; the B. 04332u. Original Image Courtesy Library of Congresshorizontal timbers seem to have been split in a rather hap-hazard fashion and were not given uniform widths or lengths. Some attempt seems to have been made to anchor the posts to the body of the parapet using cut "V" shaped branches. These anchoring branches were positioned about 1 to 1 1/2 feet above the foot of the interior slope, this would have been was too low to influence the stability of the upper reaches of the posts. This may be taken as another indication of the inexperience of the engineer or other officer assigned to direct the actual construction of the work.

B. C. 04332u. Original Image Courtesy Library of CongressThis close view also confirms that there was a distinct separation between this redan and the next battery (C) to the north. Looking through breaks in the cask traverse the sloping extremity of the next battery can clearly be seen over the top of the redan's left face parapet revetment. The parapet of the next battery can be seen rising to a level and maintaining that level up to the nearest embrasure of the battery. The heads of this work's revetment posts form a distinct and even line that marks the work's interior crest. Both of this work's two embrasures are also visible; their cheeks were given a post and horizontal timber revetment. Horizontal timbers were cut to follow the profile of the parapet's exterior slope. Since the interior of the battery can not be seen, it can not be determined whether these embrasures were arranged for heavy naval ordnance maneuvered with tackle, but the probability is that it C. 04332u. Original Image Courtesy Library of Congresswas. Unlike the redans to its right (south) and left (north) this battery's interior crest was not raised higher than that of the adjoining line of parapet. This was probably due to the fact that this was a small work intended strictly as cover for its artillery armament and did not include any substantial length of banquette for a musket armament.

III. Northeast Redan (D)

D. 03910u. Original Image Courtesy Library of CongressThe field work at the northeast angle of the interior line of works (marked D on the map) exhibits a number of peculiarities that demand attention. Its trace was very similar to the first work north of the railroad: it was a redan with two distinct faces that included a curvilinear raised battery at the salient angle. The parapet of the right (south) face can be seen above the barrel traverse of the first redan north of the railroad (B); it appears to be a rather normal parapet prepared for a musket armament. This line of parapet joins the raised mound of the salient battery at the point where an exterior revetment sustaining the lower portions of the battery's exterior slope at a vertical begins. This revetment, which is supported by 5 long timber shores, follows the outline of the battery. As with the previous redan, the parapet of the battery is raised well above the height of the infantry parapet forming the redan's near face. A single embrasure can clearly be seen in battery parapet with its cheeks sustained by a post and timber revetment. A tall post and timber revetment seems to cover the rear of the salient battery. Another line of revetment can be followed from the salient battery toward the left where it become very indistinct and mixes with the background.

D. 04332u. Original Image Courtesy Library of CongressTo examine this work in a bit more detail it will be necessary to call the closer view offered by the second photograph (03442u) into service. This photograph shows that the redan was covered by a continuous ditch that appears to have had somewhat greater dimensions than the ditches in front of the other redan to the south (B). The exterior slope of the infantry parapet falls onto a very slight berm, no more than a few inches wide, just above the crest of the scarp. But this berm widens out below the exterior slope revetment of the salient battery parapet and takes on a slope of varying degrees until it reaches within a few yards of the point where the parapet's curve takes it out of the camera's view where this berm becomes quite flat. The infantry parapet was D. 04332u. Original Image Courtesy Library of Congressnot on a level, but rises gently from its extremity (to the left, above the small two gun battery (C)) to the point where it joins the raised salient battery. It is possible to make out the tops of a few of the parapet's revetment posts that mark the line of the interior crest. The two dimensional nature of the camera's point of view does not allow a very good impression of the exterior lines of the parapet; the superior slope seems to fall onto the exterior slope without assistance of a strong line of demarcation.

D. 04332u. Original Image Courtesy Library of CongressThe transitional area where the infantry parapet joins the raised parapet of the salient battery is influenced by the presence of the exterior revetment of the battery's exterior slope, which deforms the last few feet of the infantry parapet's exterior slope. The gentleman holding the water bucket unfortunately blocked the camera's view of the actual transition between the unrevetted and revetted exterior slopes. Above this, the transition between the infantry parapet's superior slope and raised battery parapet is quite abrupt with the end of the battery parapet almost double the height of the infantry parapet and sustained by a post and timber revetment. This revetment appears to have a vertical angle and seems to be much too rectangular to represent the actual profile of the battery parapet; the exterior end of the revetment actually casts a shadow onto the upper portion of the exterior slope next to it.

From this end revetment the battery parapet rises in height as it approaches the revetment of the embrasure, it seems to take a definite level and keep it beyond the embrasure. The embrasure itself is really quite interesting; D. 04332u. Original Image Courtesy Library of Congressthe revetment does follow the parapet profile fairly closely, even if the parapet appears to follow a consistent slope from the interior crest without any strong lines separating the superior from the exterior slope. Enough of the interior of the embrasure is visible to determine that the sole was appropriately sloped downward from the mouth on the interior slope of the parapet to its defining exterior line along the top of the exterior slope's revetment.

Wood exposed to artillery fire was not generally not considered a good thing, it therefore seems a bit odd that this field work would include a very substantial post and timber revetment along the lower portion of its exterior slope. This revetment would have increased the difficulty assaulting troops would experience getting into the work over the battery parapet, especially when subject to flanking fire provided by the collateral two gun battery (C); but the far side of the parapet that can not be seen by the camera was not as well flanked. The revetment combined with the berm that lead to the joint between the raised battery parapet and lower infantry parapet would have provided a covered passageway for assaulting troops to easily reach the top of D. 04332u. Original Image Courtesy Library of Congressthe infantry parapet and rear of the battery. The severe consequences of the this revetment's destruction by distant artillery prior to an assault can well be imagined, that is, the parapet would probably slide into the ditch. This would both steal the defenders' cover and provide assaulting troops with a ramp across the work's ditch. It seems doubtful that any competent engineer would design this dangerous feature into a field; more than likely it was yet another result of inexperience either by way of an initial error in laying out the lines of the work on the ground or a mistaken estimate of the soil's native capacity for holding a slope under pressure of the mass of the parapet. This last point seems to be confirmed by the apparent sliding and rotation of the revetment posts that was checked by the addition of the five long timber shores that stretched from the counterscarp of the ditch to the upper portions of the posts. Given the sharp slope of the berm at the near end of the revetment it may also be assumed that the revetment was constructed after the parapet was completed specifically to prevent the weight of the parapet from crushing the scarp and carrying the protective covering mass into the ditch.

The crest of the scarp can be seen in silhouette where the battery's curve turns it out of the camera's view. It is notable that the scarp appears to have an almost vertical angle at the top before taking a sharp slope down to the bottom of the ditch. This tends to confirm that the scarp was not given an angle sufficient to sustain the weight of the parapet. The bottom of the ditch appears to vary in depth from 6 to 8 or more feet, but it also seems to lack a distinct counterscarp in the section visible in the photograph. In fact, ground around the work D. 04332u. Original Image Courtesy Library of Congressseems to roll gently into the ditch until it meets the foot of the scarp where it takes on a very sharp angle. This would have made it fairly easy for assaulting troops to reach the foot of the scarp, but once there, assuming that the exterior slope revetment was still intact, they would encounter a very serious series of obstacles. They would be compelled to scale the scarp to mount the berm, then climb up the revetment to get onto the upper reaches of the exterior slope, and only then could they try to go over the superior slope or enter the work through the embrasures. All of this while under a flanking and reverse fire from the other works composing the east front of defense. But as has already been pointed out, there would be no reason to go over the battery parapet when an easier route around it had been built into the work.

D. 04332u. Original Image Courtesy Library of CongressUnlike the interior slope of the two works previously discussed, this battery did not have a split revetment; judging from the portion of the interior slope visible to the camera, this work had a complete post and timber revetment. This revetment had about 13 horizontal timber laid one on top of the other that were retained in place by 7 rather tall posts. At least one of the revetment posts had to be sustained by a timber shore. The terre-plain of the raised battery was accessed by a ramp, which can be seen just in front of the sentry box and table. This ramp, along with other details of the interior of this battery can be seen in yet D. 01084u. Original Image Courtesy Library of Congressanother photograph (01084u) made in July, 1862 after the series work of deterioration and destruction by scavenging had been well underway for quite some time.

This photograph show the rear of the redan's right face infantry parapet and the interior of the raised battery. The battery terre-plein was raised to the level of the superior slope of the infantry parapet; two of the three embrasures can be seen, the third being hidden by the camera's angle of view. As with the other batteries, this one includes tackle posts for maneuvering naval ordnance mounted on ship's trucks. Many small branches and twigs can be seen sticking out of the interior slope of the battery D. 01084u. OriginalImage Courtesy Library of Congressparapet, indicating that this parapet was also constructed by combining the earth with small branches. The effectiveness of this combination is obvious since the interior slopes retain almost vertical angles even though the post and timber revetment has disappeared. Looking toward the exterior slope of the parapet, the revetment is still present, but the shores seem to be gone, leaving the revetment posts to continue the process of destruction by vertical rotation.

August, 2004

Modified August 19, 2004

Modified April 20, 2005