My Cart

Frames of Destruction

During my time at Mariners’ I have frequently been intrigued by an odd looking object in one of our storage areas but time wasn’t always available to learn more about it. That recently changed for one object when I spent several months researching the history behind a piece that has always intrigued me—a large, bent, barbed, piece of iron–the spear of a chevaux-de-frise.

Iron spear from a chevaux-de-frise for river use (Accession#1944.04.01/A37)

For those of you thinking “chev-o-de-what?”, chevaux-de-frise are those long lines of angled sharpened posts you sometimes see on historic battlefields. They were used as obstructions to prevent cavalry from overrunning a defensive line. For those of you who already know what they are and are asking yourself “how can one of those defend a river?”, that’s where the interesting story lies!

The transformation of chevaux-de-frise from a land-based defensive weapon into a maritime one occurred during the American Revolution. It was one of many defensive barriers deployed in rivers near major cities to try and protect them from the overwhelming might of the British Royal Navy.

Civilian diver Norman Lynch aboard the derrick Babcock examining a chevaux-de-frise spear recovered by Army engineers near Billingsport. From The Morning Post, Camden, New Jersey, March 4, 1944, page 3.

Our point was recovered on March 3, 1944 by US Army engineers conducting clearance and dredging work off Billingsport, New Jersey. This means our spear was part of the most successful application of marine chevaux-de-frise, the defense of the Delaware River and Philadelphia (which was serving as our new nation’s capital).1 While the obstructions didn’t prevent Philadelphia from being captured, fear of them led the British to change the course of their campaign and definitely slowed them down as they worked to seize control of the river.

An East Prospect of the City of Philadelphia; taken by George Heap from the Jersey Shore, under the direction of Nicholas Scull, Surveyor General of the Province of Pennsylvania. 1768 (Accession#1936.34.01/LE187)

The story starts on June 30, 1775, when Pennsylvania established a Committee of Safety to organize the colony’s defenses. Within a week, the committee had examined the Delaware River and the existing defenses around Red Bank and Billingsport. A sub-committee called the “Committee for the Construction of Boats and Machines” had also been established. Of the five ideas suggested for the defense of the river the committee accepted the plan developed by Philadelphia carpenter Robert Smith, who had also offered to oversee the construction of the “machines” without payment. Within a few weeks Smith had also designed a “machine for lowering and raising ballast into and out of the chivaux de frise2 to be sunk in the river.” This is the first time the “machines” were identified as a form of chevaux-de-frise.

Unfortunately, no contemporary plans of the structures exist, but descriptions from letters, the minutes of the safety committee, information gathered by the men tasked with their removal in 1784, and recovered pieces like ours give us a good idea of how they were constructed. The base of the typical marine chevaux-de-frise placed in the Delaware River was a thirty-foot square open-top box3 constructed of huge4 logs and lined with two-inch thick pine planks. Two or three5 17 to 20-inch diameter poles tipped with large iron spikes were placed diagonally in the box. The length of the poles/spikes varied to accommodate the depth of the water and ensure the spears never had more than six feet of water over them at low tide.

This is my own adaptation of the chevaux-de-frise used in the Delaware. The original drawing was published on a map titled “The Course of the Delaware River from Philadelphia to Chester…” by William Faden. That image was much more representational of the chevaux-de-frise used in New York and it had a fatal flaw! It showed the spikes sticking above the water at low tide!
Example of a cheval-de-frise spear recovered from the Delaware River
Example of a cheval-de-frise recovered from the Delaware River

Once constructed, the chevaux-de-frise frames were floated into position and anchors were used to lower the empty boxes to the riverbed and hold them in place while Robert Smith’s “ballasting machine” filled the box with somewhere between fifteen and fifty tons of rock.6 The individual frames were sunk about thirty feet apart7 with their spears pointing downstream at the perfect angle to pierce the bottom of any wooden vessel that encountered it. The frames were then chained together to prevent them from being canted to dump out the ballast or lifted and moved out of the way.

There is a real dearth of artwork showing the chevaux-de-frise and the havoc they caused so I recruited a few young artists I know to help illustrate my post. This image is courtesy of Carolyn W.

Over the next two years more than sixty-seven8 chevaux-de-frise frames were deployed in two tiers in the Delaware River below Philadelphia. A lower band of about twenty-four frames were sunk in two rows across the narrow channel separating Billingsport, New Jersey and March Island (also called Billings Island). An upper band of about forty-three frames were sunk in four groups in the main and secondary channels between Hog and Mud (or Fort) Island and the Jersey flats near Red Bank. The two ranges of chevaux-de-frise were guarded by land-based forts or redoubts and armed galleys and floating batteries operated by the Pennsylvania State Navy. As it turned out, the Americans were right to worry about the need to defend the Delaware River.

Cropped view of the map titled “A Plan of Delawar [sic] River from Chester to Philadelphia…” from Volume 4 of the Atlantic Neptune, showing the two tiers of chevaux-de-frise. (Call#VK985 .A88 Rare OO)

On March 3, 1777 Lt. General Sir William Howe and his brother Vice Admiral Richard, Lord Howe were given permission to launch a campaign against Philadelphia. Both men hoped seizing the new nation’s capital would bring a quick end to the war. Over the next five months the brothers assembled a fleet of more than 250 vessels to carry seventeen thousand troops, artillerymen, ammunition and provisions south. By July 23, 1777 the fleet was at sea but the Americans were still unsure of its exact destination.

General William Howe, from Volume 1 of History of the War with America, France, Spain and Holland Commencing in 1775 and ending in 1783. By John Andrews. Published in London in 1785. (Call#E208 .A56 Rare)
The Right Hon.ble Richard Lord Howe. Commander in Chief of his Majesty’s Fleets in America. Published by John Morris, London 1778. (Accession#1975.27.01)
Lord Howe’s Fleet sailing from Sandy Hook, July 1777. Watercolor attributed to Irwin John Bevan. (Accession#1940.823.01/QW208)

When the fleet arrived at the Delaware capes it was met by Captain Andrew Hamond in the 44-gun frigate HMS Roebuck. Over the course of the eighteen months Roebuck had operated in the area Hamond had developed a healthy respect for the American defenses and for the bay and river whose swift and strong tides, constantly shifting sand banks and narrow and shallow channels made navigation extremely difficult. Roebuck and its consorts had suffered frequent groundings during their time in the bay. Recognizing the difficult navigation, the high probability of having ships sunk by hidden booby traps, and suffering bombardments by an enemy that couldn’t easily be fought, General Howe reconsidered his route of attack. Rather than disembarking his troops in the Delaware, the fleet sailed to Chesapeake Bay and on August 25th landed the army at the Elk River in Maryland. A month later, after a campaign that included one of the largest battles of the Revolutionary War, the Battle of Brandywine Creek, the British marched into Philadelphia uncontested.

Light frigates of Lord Howe’s fleet anchored off Chester, Pennsylvania in September of 1777. Watercolor attributed to Irwin John Bevan (Accession#1940.806.01/QW209)

With the city under British control the Howes realized they had a serious problem–they had no hope of holding Philadelphia if they didn’t have control of the Delaware River. Yes, British ships could land cargoes at Chester, Pennsylvania and transport them by wagon up to Philadelphia, but the route was vulnerable to attack and just plain inefficient. The easiest, safest and most expedient way to get provisions and munitions to the occupying forces and remaining inhabitants in Philadelphia was by water. Thus, the Howes were determined to break the American blockade of the river. It ended up being a long, difficult and costly campaign for the British, one that involved one of the heaviest artillery bombardments ever seen on American soil.

As we’ve seen, the American defenses of the Delaware were strong. The most formidable obstacles were the tiers of chevaux-de-frise which by late 1775 had already claimed at least three victims. On November 27, 1775 Caesar Rodney wrote “while poor David Beveridge was at Lewis his brigg came in loaded with sugars, molasses and coffee. She has since run upon the chevaux-de-frise and immediately sunk in five-fathom-water, no part insured. One or two river boats have struck and been sunk by them the moment they touched. I mention this that you may give strict orders to your people to be careful in passing them with the new schooner, for if she touches them she is shorely lost, with all her cargo.”

Also formidable was the flotilla of brigs (seven), sloops, schooners, xebecs, galleys (thirteen) half-galleys (twenty-six), floating batteries (two), fire ships and rafts (seventeen), and other armed vessels (eight) operated by the Continental and Pennsylvania State navies. The low profile of the galleys and batteries made them extremely hard to hit while their shallow draft and maneuverability allowed them to swoop in, attack and quickly leave. By contrast, the larger, deeper draught, wind driven Royal Navy ships couldn’t easily maneuver amidst the Delaware’s constricted channels and constantly shifting sandbanks. The American fire ships and rafts weren’t as much of a threat, but they certainly made the British sit up and take notice when one of the flaming vessels drifted towards them on the evening tide.

HMS Phoenix grappled by an American fire ship in the Hudson River in August 1776. The ship managed to fight its way free with minimal damage but was forced to drop back downriver to a new anchorage, which is what repeatedly happened to the British in the Delaware River. (Accession#1945.420.01/LP2951)

Less formidable were the land and naval defenses protecting the tiers of chevaux-de-frise. Unfortunately for the Americans the forts were ill-conceived, ill-prepared and critically undermanned. The one exception was Fort Mifflin which sat on a muddy island in the middle of the Delaware. It was fairly safe from being stormed by troops but vulnerable to artillery bombardments. In hindsight, if Howe had landed his army south of the chevaux-de-frise in July he probably would have easily overcome the American defenses. Instead, he gave the Americans another month or so to prepare which allowed them to successfully defend the river for nearly eight weeks.

“Plan and sections of the redoubt at Billingsfort…” The initial fort at Billingsport had been overambitious and there weren’t forces available to garrison it. By 1777 it had been contracted to a small redoubt protected by a row of abatis (sharpened branches of trees laid in a row facing the enemy). Library of Congress Geography and Map Division (Call#G3814.P32:2B5S3 1777 .P5)

As I mentioned earlier, the first line of defense was the lower tier of chevaux-de-frise which was protected by a small redoubt at Billingsport, New Jersey and American galleys and floating batteries. Aware that the installation at Billingsport was seriously undermanned and vulnerable to attack from the land Howe sent 1500 troops to take the fort forcing the 100 or so American defenders to evacuate. With the redoubt in British hands Hamond and a small squadron of British ships began working to create an opening through the lower chevaux-de-frise about “half a cable’s length” (about 300 or 400 feet) from the Billingsport shoreline.

Their task wasn’t a walk in the park. The Americans had done their job well and the individual chevaux-de-frise frames were extremely difficult to move or dismantle. Their first attempts began on October 4, 1777 when HMS Pearl reported “broke our hawsers in purchasing them” (a hawser is generally 5-inches in diameter or larger). On the 13th HMS Liverpool’s logbook reports “Brot [brought] the hawser to the capston, ½ past carrd 1 of ye hawsers away [yet another hawser broken!], unbent the stream cable & swept the Cheveaux de frize with it, lost our stream anchor amongst the cheveaux de frize. At 1pm got the stream cable fast to the chevaux de frize, bent the messenger to it, & hove on all, but did not move them…” The same day HMS Vigilant managed to get stuck on one of the chevaux-de-frise. The British broke two more hawsers trying to heave the vessel off and around 10:00 PM managed to pull it free– apparently without damage.

HMS Roebuck attempting to remove the lower chevaux-de-frise in early October 1777. In the background you can see the Pennsylvania State galleys attacking the small boats working to remove the impediments. Watercolor attributed to Irwin John Bevan. (Accession#1940.796.01/QW216)

To add insult to injury the British had to work while under continual bombardment by American naval forces. The heated firing repeatedly forced the British to withdraw their ships and men to safer locations downriver–well, somewhat safer, as their logbook’s report ships continually running aground! It would take until October 14th before HMS Liverpool reported “at 4pm we had a passage 17 fathoms wide” (about 100 feet). It must not have been wide enough because the British continued working and didn’t manage to warp (i.e. pull their ships through on anchor cables) the first ships through the barrier until October 20th.

Detail from the William Faden map “The course of Delaware River from Philadelphia to Chester, exhibiting the several works erected by the rebels to defend its passage, with the attacks made upon them by His Majesty’s land & sea forces” showing the opening in the lower tier of chevaux-de-frise. Library of Congress Geography and Map Division. (Call#G3792.D44S3 1777 .F3)

The clock, however, was ticking for the British. If they didn’t gain control of the river before winter set in, the British occupiers and the citizens of Philadelphia would face starvation. To break the stalemate, Howe ordered a simultaneous attack by land and river on Fort Mercer on the Jersey shoreline at Red Bank. Fort Mercer was the supply and manpower lifeline for Fort Mifflin. British ships began warping up the river on the afternoon of October 21st to support an attack by a Hessian brigade 1200 strong. The October 22nd attack was a total disaster! Not only were two British ships destroyed (one possibly because it ran onto a chevaux-de-frise9), but the Hessians were repulsed with heavy casualties by the few hundred Americans stationed within the fort.

Representation of the destruction of the 64-gun ship HMS Augusta and the 16-gun sloop HMS Merlin. In reality Augusta caught fire in the morning and blew up around noon while Merlin was intentionally set on fire and blew up around three in the afternoon. Watercolor attributed to Irwin John Bevan. (Accession#1940.797.01/QW217)

Over the next three weeks both sides feverishly worked to prepare for the siege of Fort Mifflin everyone saw coming. The Americans reworked, strengthened and expanded the defenses of the fort. They also established several temporary shore-based batteries above Billingsport. At the same time the British established numerous batteries near Fort Mifflin and at night ferried supplies in flat-bottomed boats up the western channel to Philadelphia. Skirmishes and bombardments between the ships, the land-based forts and batteries, and troops in the field occurred daily as each side tried to disrupt the other’s preparations.

At the same time HMS Liverpool continued working to widen the gap in the lower chevaux-de-frise–again without much success. Reports from Liverpool’s logbook are a running commentary of failures: on October 25th “swept the cheveaux de frise with 2 hawsers in order to widen the passage, lost a boats grapnell”; on the 30th tried “to overset the cheveaux de frise, but without success, having broke an eight and nine inch hawser”; on November 1st “at noon got the purchase blocks on the hawzer and hove on all. At 2PM carrd [carried] away the strap of the purchase blocks.” The same day Liverpool found herself hung by her quarter on the chevaux-de-frise. The grounding severely damaged the ship’s false keel which required significant repairs the following month. Despite the damage the crew resolutely continued working and on November 4th they reported “heaving in the cable made fast to the cheveaux de frize, found we draw’d them nearer to us but did not widen the passage already made.” Can’t you just imagine the cursing that was probably going on by this point?

Representation of the action between ships on November 10, 1777. Despite what the artist depicts, the galleys of the Pennsylvania navy were not armed to fire broadsides. Watercolor attributed to Irwin John Bevan. (Accession#1940.799.01/QW218)

At dawn on November 10th the British began a massive bombardment of Fort Mifflin. Thousands upon thousands of shells rained down on the fort all day, every day and the Americans effectively returned the fire with every gun that remained available. Each night, the Americans stubbornly repaired the damage. It took five days before the Americans ran out of ammunition and the damage became so extensive it just couldn’t be easily repaired. Overnight on November 15th, the Americans quietly set fire to the fort and evacuated.

Detail from “The Course of the Delaware River from Philadelphia to Chester with the several forts and stackadoes raised by the Americans and the attacks made by His Majesty’s Land and Sea Forces” by William Faden. Library of Congress (Call#G1201.S3 F2 1845)
“A Plan of Fort Mifflin on Mud Island, with the Attacks made by the Kings Troops and Vessels.” Detail of map by William Faden. Library of Congress (Call#G1201.S3 F2 1845)

Although Fort Island was now in British hands they still didn’t have free passage of the river. The Americans still held Fort Mercer and the river was still filled with our warships. These active defenses made working to create passages through the chevaux-de-frise hazardous. If they did find a way through, the British transports would be at the mercy of the American warships. It took the threat of 6,000 British troops marching on Red Bank to convince the Americans to abandon Fort Mercer and move as many ships as possible up the Delaware beyond Philadelphia where the deeper draft British ships couldn’t follow.

The American ships that weren’t able to escape up the Delaware were burned to prevent them from falling into British hands. Watercolor attributed to Irwin John Bevan. (Accession#1940.800.01/QW219)

So now the British finally had control of the Delaware…or did they? Winter had arrived and the passage was still very narrow but there was nothing they could do until warmer weather arrived. Thus the chevaux-de-frise continued to claim victims:

  • December 1st the naval store ship Juliana and a brigantine from Glasgow with bale goods were sunk (Juliana was raised about twelve days later);
  • December 9th the ordnance transport Rebecca was sunk;
  • January 7th, 1778 the Crawford was sunk;
  • In February the ships Mary Hannah and Two Sisters, which sadly took their entire crew, were sunk;
  • On March 30 high winds caused a larger ship to crash into the sloop Brilliant and push her onto a chevaux-de-frise. It sank.
Artwork by Evan W.

The final “ship stabbing” I was able to find occurred on May 16th when the Pennsylvania Ledger reported the 200-ton ship Albion “in coming up the river…struck upon the upper cheveaux de frize, from which he got off, but has received considerable damage.” The damage must have been significant because on June 27th Albion sank.

When spring arrived the British ships went back to work trying to widen a passage through the chevaux-de-frise. They must not have made much progress because Captain James Parker reported on May 3, 1778 that sheer hulks10 were being constructed to try and help remove the obstructions. I don’t know if the vessels were ever used because a little over a month later the British abandoned Philadelphia. As it turned out, seizing control of Philadelphia and the river had absolutely no effect on the progress of the war.

Sheer hulk shown in an engraving titled “View of His Majesty’s Dock Yard at Chatham in the County of Kent, on the River Medway.” (Accession#: 1936.482.01/LE462)

By 1783 the buildup of sediment around the chevaux-de-frise had essentially formed an island in the middle of the river and had seriously constricted the width and depth of the channel. Ships were still being damaged by the structures so the merchants of Philadelphia petitioned to have the chevaux-de-frise removed.11

Advertisement requesting proposals for the removal or destruction of the chevaux-de-frise. From The Freeman’s Journal or The North-American Intelligencer, Philadelphia, Pennsylvania, April 21, 1784, page 2.

A contract to remove the obstructions was awarded to Levi Hollingsworth and Arthur Donaldson in May of 1784. In June, the Pennsylvania Packet reported that Donaldson had constructed a “vast apparatus” with which he had succeeded on the first attempt to raise “the largest chevaux de frise, that was sunk in the deepest water” (not sure how they knew this was the largest and deepest, obviously sensationalizing a news report to expand readership isn’t a new thing!). One does wonder, however, if his “vast apparatus” was some form of sheer hulk. Whatever it was, it worked. By September Hollingsworth and Donaldson had removed sixty-four of the sixty-seven reported chevaux-de-frise frames found from the river.

Plat showing the lower range of chevaux-de-frise frames. Those marked “A” were additional frames found during the removal project. The frames marked “B” were those “destroyed by the British or other causes” and were left behind. Pennsylvania Archives, Volume X.

Donaldson and Hollingsworth indicated they did not remove the four frames that had been substantially destroyed by the British in 1777-1778 because they didn’t pose a threat to navigation. This statement immediately made me sit up and take notice! Most of the cheval-de-frise that have been recovered over the years all seem to be mostly intact poles/spears. Ours, on the other hand, is only the iron spear and it is weirdly bent. It was also recovered in the area where the British were attempting to force an opening. This certainly makes me wonder if our spear belonged to one of the frames destroyed by the British in 1777–which is pretty darn cool!

If you want to learn more about the defense of the Delaware River during the American Revolution I highly recommend the book “The Pennsylvania State Navy, 1775-1781, The Defense of the Delaware” by John W. Jackson. It’s a well researched and interesting publication. I’ve already read it three times!


1 Chevaux-de-frise were also employed in several places in the Hudson River but didn’t perform as well because of the depth of the river and incomplete nature of the obstructions, but they did make life difficult for the British! On December 24, 1777 the frigate HMS Mercury ran upon the chevaux-de-frise off Fort Knyphausen and sank with the loss of three drunk marines. Chevaux-de-frise were also used to defend Charleston and Boston harbors.

2 You gotta love 18th century spelling. I found chevaux-de-frise spelled eleven different ways while I was researching: the correct spelling, chivaux, chiveaux, cheveaux, chivoux, chieavau, chive de freax, chevack de friez, chevause de frieze, chevee de frizes, and, my favorite, sevardefrize (all one word).

3 Many give the dimensions as 40 x 60 feet, but that is the dimension of the chevaux-de-frise built to protect the Hudson River, which needed to be much larger.

4 In a letter from the Committee of Safety to Colonel Stephen Moylan in New York the committee states: It takes about 25 or 30 logs from 40 to 65 feet in length, from 12 to 26 inches thick in the butts for each of the chevaux de freize sunk in our river.”

5 Only the marine chevaux-de-frise in New York had four cheval-de-frise in each box.

6 The reported weight of ballast used varies, some early reports say fifteen to twenty tons while Captain Andrew Hamond of HMS Roebuck reported forty to fifty tons. The 1944 Morning Post article reported that boulders weighing as much as 42 tons had been recovered in the area where our spear was found which certainly makes it seem as though there was more ballast holding the boxes down than people realized!

7 The committee specified the frames were to be sunk with sixty feet between the frame centers which, when calculating their 30’ dimension, puts them roughly 30’ apart.

8 The exact number built isn’t known. Sixty-three were removed in 1784, three or four others had been destroyed by the British and their remains were left behind because they weren’t a navigation hazard, and a number of remnants have been found over the years. Some were built but never deployed.

9 British captain John Montresor reported in his journal that the sloop Merlin struck upon the chevaux-de-frise near the Jersey shore. When the British couldn’t get her off they set her on fire to prevent her from falling into American hands.

10 Sheer hulks are vessels with strengthened masts capable of supporting a purchase and mechanical power to raise or remove heavy objects. You typically see the near dockyards being used to raise and lower masts. They’re sort of like a floating crane.

11 On June 10, 1783, The Pennsylvania Packet reported that the ship Achilles struck on the chevaux de frise on her passage up the river and sustained enough damage to put her in danger of sinking. She was saved by “the timely assistance of a number of inhabitants of the southern liberties.”

Scroll to Top