Let's Play "Name That Truss!"

One of the major goals of this website is to categorize bridges by a wide variety of things: their design, status, builders, location, and more. For truss bridges, the design types (Pratt, Howe, Warren, etc.) are fairly well established, but not completely. Sources don't always agree about the design for a particular bridge. Engineers liked to experiment, producing oddball designs that don't quite fit within the standard categories. Other times, the design might be fairly common, but lies within a grey area that makes it hard to classify.

Here are some examples that I've pulled from this site. How would you classify these bridges?

Queenpost vs. Pratt

Traditionally, a Queenpost truss has three panels, with the center panel lacking any diagonals. This rudimentary design can be seen in this Pennsylvania covered bridge:

(Photo by Jack Schmidt)

The lack of diagonals makes for a somewhat flimsy truss. Thus, it makes sense to add diagonals, such as this Wisconsin bridge that features wooden diagonals to go with iron verticals.

(HAER photo)

From there, it's not a stretch to build an all-metal bridge with diagonals.

(Photo by Satolli Glassmeyer)

However, at what point does the Queenpost truss turn into a 3-panel Pratt pony truss? For the bridge in Wisconsin (the second photo), the HAER documentation clearly calls it a Queenpost. But sources are unclear about the third bridge. I've seen documentation describe this kind of bridge as both a Queenpost and as a small Pratt.

Miller & Borcherding truss

Some covered bridges with a Queenpost truss also include additional braces, such as this Pennsylvania bridge:

(Photo by Jack Schmidt)

What if you take this design and turn it into an uncovered metal bridge? You get something like this:

A Missouri company, Miller & Borcherding, developed this truss as their specialty. Is it a Queenpost? The Missouri Historic Bridge Inventory calls these bridges a "hybrid Pratt/Warren design", a real mouthful. I've been classifying this design under its own category as a Miller & Borcherding truss. However, would it make more sense to call it a Queenpost variant? Or a Warren variant?

Bonus question: Let's say you took this design, turned it upside down, and treated it as a deck truss. What the heck would you call this private bridge I stumbled across in Missouri?

4-panel trusses

Is this Arkansas bridge a Pratt or a Warren truss?

(Photo by Wayne Kizziar)

With only four panels, it's hard to tell. The diagonals form both a "V" (Pratt) and a "W" (Warren) profile. I've classified these as Pratts, but the case could be made for Warren trusses.

5-panel trusses

For through trusses, most sources draw a distinction between Parker trusses and Camelback trusses. A Camelback is a Parker with exactly five slopes, usually with a flat top that extends across multiple panels.

For pony trusses, however, the distinction is less clear. I've seen these bridges described as both Parker and Camelback trusses.

(Photo by Wayne Kizziar)

Since the Camelback is a special case of the Parker, it is technically correct to use either term. Is it worthwhile to make a distinction? Or should the Camelback term be limited to just through trusses?

Baltimore vs. Warren trusses

Take a look at this bridge in Arkansas:

(Photo from AHTD)

The Arkanas historic bridge inventory classifies this as a Baltimore truss. With the flat top and subdivided panels, this makes sense. But look closer: the diagonals form an "A" in the center. Across all four panels, the main diagonals alternate: the hallmark of a Warren, and not a Baltimore (Pratt), truss. So is this bridge really a Warren truss with subdivided panels?

Subdivided Warren trusses

Speaking of subdivided Warren trusses, I've seen the term used for two completely different kinds of trusses. One features additional members within each panel, similar to the way that a Parker truss is turned into a Pennsylvania truss:

The other style takes an ordinary Warren truss and adds verticals at all panel points:

(Photo by Calvin Sneed)

Which is right?

Warren vs. Lattice pony trusses

Here's another puzzle involving Warren trusses. What exactly is this Missouri bridge?

I've seen these described as "double-intersection Warren pony trusses", "lattice pony trusses", and as "lattice girders." For such a small bridge, it's hard to tell whether the "truss" railings are load-bearing, or whether the stringers bear the load with the railings merely serving as ornaments. Or both. For that matter, is this even a truss bridge?

Kingpost trusses

A classic Kingpost pony truss is easy to spot:

(Photo by Charles Robinson)

The Wadell "A" through truss is also easy to spot, although it is extremely rare:

(Photo by Joshua Collins)

Both what if you take a Waddell "A" bridge and reconfigure it as a pony truss? Does it become just a Kingpost? Or is it sufficiently different from a Kingpost to call it something else (like "Subdivided Kingpost" or "Waddell pony truss")?

(Photo by Robert Elder)

Now take that pony truss and attach a single overhead brace. What would you call this Indiana bridge?

(Photo by Robert Stephenson)

Better yet, say you took a Kingpost truss but added Warren-style diagonal girders. What the heck would you call this Ohio bridge?

(Photo by Gregory S. Hamilton)


K-trusses, limited to just a handful of states, never gained widespread acceptance, perhaps because they came on the scene when truss bridges were falling out of fashion. A full K-truss, with diagonals forming the "K" shape at all of the panels (except the hips), is easy to classify:

(Photo by David Backlin)

But what about a bridge with the "K" diagonals within only a fraction of the panels? Should this be called a K-truss as well, or should it be classified as a hybrid with a Parker truss?

Here's an example from Tennessee with the "K" at the two center panels only, flanked by ordinary Pratt-style panels on the outsides:

(Photo by James Adorno)

And here's an example from Oklahoma with the "K" at the outside panels and an ordinary "X" at the two center panels:

(Photo by Gene McCluney)

How should these be treated? Does it even matter?

Head scratchers

If you thought some of the examples above were tricky, here are the real puzzlers. Consider the English Center Bridge in Pennsylvania:

(HAER photo)

This posted quite a challenge to the HAER researchers. The initial HAER report described it as an eyebar-chain suspension bridge. Upon further review the following year, a more detailed report concluded, with some reservations, that it was really a "two-hinged inverted trussed arch" and not a suspension bridge at all.

OK, fine. But how do I classify that exactly? Is it better to treat it as a truss or as an arch? If it is a truss, would it be a Pratt variant? Or something entirely different?

Meanwhile, the Asylum Bridge in Kansas is enough to make an engineer go insane.

(Photo by Robert Elder)

It sure looks like a cantilevered through truss. However, the consensus is that the bridge is not cantilevered at all, but sports a "reverse Parker" main span. I'm still having trouble wrapping my head around this one. To me, a reverse Parker should look like this:

(HAER photo)

This is part of one of the many viaducts approaching the Hell Gate Bridge in New York City. These three spans are typical Parker trusses flipped over to create a deck truss. Are these also "reverse Parker" spans? Or is it better to call them "inverse Parker" spans? Before you answer, consider the Victorian Footbridge in St. Louis, Missouri:

While this looks like a suspension bridge at first glance, it probably functions as a truss. If it is a truss bridge, does it quality as a reverse Parker?

Feel free to discuss any of these questions or to ask about other mystery trusses.

Coming next: Trying to make sense of the many different kinds of portal braces used on through trusses.

Comments  (21)

Let's Play "Name That Truss!"
Posted February 16, 2010, by Melissa Kernea (rebellovers30721 [at] yahoo [dot] com)

Knows that since Eddie Douthitt and I have been doing the bridge site here we have come across a few that we have had to step back and play the "name that truss" game that it had led to long talks, head scratching and an all around good time...just wanted to say thanks to all who do such hard work and interesting conversation on this site and we hope to keep working on this site as long as we can....

Let's Play "Name That Truss!"
Posted February 5, 2010, by Robert Elder (robertelder1 [at] gmail [dot] com)

Thanks for the feedback on the Duncan Creek Bridge. Good to know that we are all in agreement on this.

Let's Play "Name That Truss!"
Posted February 4, 2010, by Nathan Holth (form3 [at] historicbridges [dot] org)

Robert, I agree, a four slope Pratt truss such as that shown below is indeed a Parker. Here is a four slope polygonal Warren:

Let's Play "Name That Truss!"
Posted February 4, 2010, by Robert Elder (robertelder1 [at] gmail [dot] com)

I am anonymous from below - with the four panel Parker through truss.

Let's Play "Name That Truss!"
Posted February 4, 2010, by Anonymous

So, if a Parker must have at least five slopes...

The general consensus suggests that this bridge is a four panel Parker through truss, which I think is reasonable.

KSHS calls it a Pratt variant, a classification with which I disagree.

Let's Play "Name That Truss!"
Posted February 3, 2010, by Wes Kinsler (webmaster [at] wkinsler [dot] com)

This is actaully a rather fun topic.

I've tried to do it on a page I put together on Oklahoma's standard plan bridges at

It is still a work in progress, and I am planning to do a similar write-up on the non-standard bridges too! Should be fun. And you will note I have drawn a distinction between camelback and Parker types in pony trusses.

There seems to be a lot of mixing/confusion between modified types and sub-diveded types, such as a modified Pratt vs. a Baltimore.

And how many know what a Collinson strut is in a truss bridge? (not a trick question!!)

Let's Play "Name That Truss!"
Posted February 1, 2010, by Nathan Holth (form3 [at] historicbridges [dot] org)

Maybe it could be applied to the overall shape of the towers, but in comparing the Golden Gate's towers overall appearance and function to the Vierendeel truss design, you should at the same time be aware that portions of the tower braces are really composed of lattice (well, double-Warren in truss speak). Its just covered up on Golden Gate. You can see the design in the Strauss patent for the Golden Gate Towers.

Let's Play "Name That Truss!"
Posted January 31, 2010, by Cliff D (clif30 [at] hotmail [dot] com)

Finally took a look at this thread and low and behold Central Louisiana's OK Allen (US 71 Red River) bridge is on display in all its K-Truss glory! I took some pictures of the K Deck trusses this afternoon and some distant shots of the through K Truss are in the background of the railroad bridge I uploaded as well.

Let's Play "Name That Truss!"
Posted January 30, 2010, by James Baughn (webmaster [at] bridgehunter [dot] com)

Almost forgot... Waddell also wrote in his book, "In respect to pony-truss spans the author has maintained for years that such structures should be ruled out entirely, and that under no circumstances is it necessary to build them, because they can be replaced by plate-girder spans by the expenditure of more money. The main objection to pony-truss bridges is that no man can tell even approximately what is the ultimate strength of their wholly or partially unsupported top chords." (Chapter 22, page 479)

Let's Play "Name That Truss!"
Posted January 30, 2010, by James Baughn (webmaster [at] bridgehunter [dot] com)

One thing to keep in mind about the Indiana Bridge Company's "High-Triangular" design: In those days, Warren trusses were often called "triangular trusses."

See Chapter 22 of J.A.L. Waddell's "Bridge engineering" book:

He gives diagrams of several kinds of Triangular and Warren trusses and explains, "The term Warren truss or Warren girder was originally applied only to a particular case of the Triangular truss in which the web triangles are all equilateral; but later writers generally use the name for any triangular truss." He added, "As there is no special advantage in making web triangles equilateral, there does not appear to be any good /raison d'etre/ for the use of the true Warren type."

Let's Play "Name That Truss!"
Posted January 30, 2010, by Robert Elder (robertelder1 [at] gmail [dot] com)

Thanks for the Pegram truss links and discussion.

As for the Bedstead design from another thread, here is one of the most bizarre trusses I have ever discovered.

Initially, it appears to be a Parker with vertical endposts. Recent photos indicate that the endposts extended through the floor, which would classify it as a Bedstead Parker through truss.

The extension of the enposts through the floor may have been a recent modification however. This is not visible in older photographs because they lower portion of the endposts were either encased in concrete, or did not exist at all.

Let's Play "Name That Truss!"
Posted January 29, 2010, by Anthony Dillon (spansaver [at] hotmail [dot] com)

Thank you HP for the information on the Vierendeel truss and how it works. It is fortunate that we have a few remaining examples of this unique type. And thank you Nathan for the links on this and the Pegram truss.

Looking at the original drawings of the Pegram, I see that the center member is actually a true vertical. As these members get closer to the ends of the bridge they flare inward, with the degree of inclination changeing on each one. This puts me in mind of the Wernwag truss, which was very similar to a Multiple Kingpost (Burr)truss except that the verticals flared outward.

In gleaning through Mr. Pegrams patent application, it appears that he was trying to stray away from being classified with the Pratts and the Warrens. Indeed, at least to my eyes, it appears he may have succeeded. I can look at a Pegram truss one time and think of a Warren, and look at it again and a Pratt comes to mind. This truss definitely falls into that grey area between.

In comparison, Indiana Bridge Company's "High-Triangular" truss is undoubtedly a Warren truss with a polygonal upper chord and leaves less to the imagination.

James, I had actually heard that somewhere about the towers on the Golden Gate being of Vierendeel design.

Let's Play "Name That Truss!"
Posted January 29, 2010, by James Baughn (webmaster [at] bridgehunter [dot] com)

The Vierendeel design has also been incorporated into the towers of major suspension bridges, including the Waldo-Hancock Bridge and the Golden Gate Bridge.

An idle Google search also revealed a building in downtown St. Louis built with a Vierendeel framework. See the NRHP documentation:

Let's Play
Posted January 29, 2010, by The Happy Pontist (happypontist [at] googlemail [dot] com)

Sorry, don't know why I was anonymous in the previous post, some glitch.

The Vierendeel design functions as a beam or girder, as does any planar truss (i.e. as opposed to a space truss).

In a conventional I-shaped beam, simply supported at its ends, the top flange carries compression, the bottom flange tension, and the solid web carries a pattern of tension and compression at varying angles (compression vertical at the ends, and diagonally between the ends and midspan). This pattern of web stresses is what is normally thought of by engineers as "shear".

The truss is essentially an I-girder with holes cut in the solid web. The upper and lower chords carry compression and tension, as in the I-girder. The web members carry tension or compression according to their orientation and the variable position of the load. In a triangular truss, because they are isolated members, the web forces are constrained by the member orientation, rather than free to vary in response to load as in a solid web.

The Vierendeel is again an I-girder with holes cut in the solid web, just rectangular holes instead of triangular holes. The top and bottom chords therefore work as in any other truss. The verticals are generally in tension if the bridge deck is at the lower edge of the truss.

Intuitively, the "holes" in a triangulated truss do not distort much in shape, and the structure is stable even if the web members have no bending stiffness. In a Vierendeel, if the web members (or chord members) had no bending stiffness, the rectangular openings would become mechanisms and the truss would fail. The Vierendeel uses their bending stiffness to prevent the rectangles distorting into parallelograms, a phenomenon most prevalent towards the ends of the span, where the shear force and hence distortional tendency is greatest (the effect of greater shear force can also be seen in triangulated trusses, where web members are often bulkier towards the span ends).

The result is a truss which is almost always less efficient than any triangulated truss, because much of the bulk of each member is used to resist local bending distortion, rather than contributing to the overall load carrying capacity. Although some Vierendeel highway and rail bridges have been built, the type is therefore generally confined to footbridges (I've designed one or two). It is also used extensively in building frames, where there are obvious benefits to having rectangular rather than triangular openings.

Of course, there are complications, principally relating to how the various systems prevent the truss/girder from buckling: in many trusses, some of the web members are there to prevent buckling or carry local load effects rather than to contribute to the overall beam behaviour.

Let's Play
Posted January 28, 2010, by Nathan Holth (form3 [at] historicbridges [dot] org)


BRIEF writeup on the Vierendeel I consider myself a bridge historian, maybe a bridge inspector when I am having a good day, but I am not an engineer. That said, I would expect the Vierendeel to act like an arch bridge, since it has no diagonals. Perhaps someone with engineering background will speak up here. I know Eric DeLony has visited the Los Angelas ones.

As for Pegram trusses, there is a TON of history on Pegrams in this Multiple Property National Register Nomination I found recently: I know the filename says Utah, but its actually for Idaho.

and of course the original patent:

These above links will be included in a page I am making for a Pegram truss that I will release next week.

Let's Play
Posted January 28, 2010, by Anthony Dillon (spansaver [at] hotmail [dot] com)

Okay got me with that Vierendeel truss. I have to admit, that is a truss I know absolutely nothing about. I would love to know how the tension and compression is distributed in one of those spans.

I think it should be noted as well that some of the terminology we use varies from one region to the next. I have always known a 5-slope Pratt through truss as a Camelback, and have never heard it used with a pony truss. Robert Elder and myself have previously discussed the Pegram truss , and a similar type that the Indiana Bridge Company built which they referred to as a "High-triangular" truss. Comparison reveals much in common, with the major difference being cylindrical eyebars hanging vertically on the IBCo spans. I don't believe these serve any function other than to support floor beams in between every other diagonal member. The crafted members on the Pegram truss are really more of an inclined (closer to vertical) than a diagonal element. In the High triangular spans, the crafted members are in compression while the eyebars are in tension. I would figure the Pegram truss to function in the same manner. One other thing of interest on the above example of the Pegram truss is the use of counters in the 3 middle panels. Counters were commonly used in the Pratt truss and it's variations (Whipple, Parker, and Camelback), but I'm not familiar with it being used in any other type of Warren truss.

Some of the unique spans remaining out there are simply a result of experimentation, and without documentation may be hard to adequately define. It does, however, give us an enjoyable topic to ponder and discuss.

Let's Play
Posted January 28, 2010, by Nathan Holth (form3 [at] historicbridges [dot] org)

In response to Robert Elder's Comments:

-You could create your own name for these Warrens which look like Baltimores, if it helped clarify the information you were trying to convey. You might call them "Warren-Baltimore" trusses. I have several of these bridges on although I chose to lump them together with some other unclassifiable Warren variations under the category "Modified Warren" but that was just a personal choice.

-Hells Gate: Yes, I agree with this assessment. The Hell's gate approach spans are simply Parker deck truss. The truss is not inverted or reversed, its simply upside down since thats what a deck truss is.

-Camelbacks: The reason a subclassification for Parkers was created called Camelback was because in the late 1800s, bridge builders wanted to improve efficient use of materials so they looked at creating polygonal truss forms. For the sake of simplicity and consistency, they chose to build bridges with exactly five slopes... which was basically just adding two more slopes to a trapezoidal truss. This was a relatively simple change that reduced materials and/or increased strength for materials used. Putting more slopes in such as found with most Parkers became too complicated, but this one change was feasible. By the way, these camelbacks were usually just done on through trusses. That is why there is the discussion about whether a 1930s pony with five slopes should be called a camelback. For sake of clarity, I DO call these 1930s pony trusses with five slopes a Camelback, but in reality these 1930s/20s bridges were built with five slopes for other reasons like span length rather than how easy it was to make a camelback versus a Parker, so they are out of context in that sense.

Let's Play "Name That Truss!"
Posted January 28, 2010, by Robert Elder (robertelder1 [at] gmail [dot] com)

There has been some great discussion on this topic already. I am adding a few thoughts to the thread.

1. Baltimore vs. Warren. There is a truss bridge in Cherokee Co. Kansas which is identical to the Arkansas bridge.

I had classified it as a Baltimore originally, however, after reading Mr. Holth's comment, I am tempted to change it to Warren. Is there a compelling reason to create a subdivision of the Warren classification for these variations?

2. Kingpost vs. Waddell. The Kansas State Historical Society (KSHS) has classified the small truss bridge in Doniphan Co. as a Waddell Truss.

This would seem to be a reasonable classification as it follows the same configuration as the Missouri example, aside from being a pony span.

3. Hell"s Gate Bridge vs. Asylum Bridge. The earliest reference to the term "Reverse Parker" that I have personally found is from KSHS.

If there were any earlier uses of the term, I am unaware of them. Thus, I will attribute this classification to KSHS for now.

I would not apply the same term to the Hell's Gate Bridge. Those deck trusses feature a Parker configuration in which the entire truss is inverted. However, only the top chords, not the entire truss is inverted on the Asylum Bridge.

4. Victorian Footbridge.

If it is a truss, then perhaps it should be classified as an inverted arch, again similar to the English Center Bridge. Both bridges feature a top chord that appears to be an inverted arch as opposed to a segmented top chord like the Asylum Bridge. If we are going to classify the Victorian Footbridge as a Reverse Parker, we should note that it has vertical endposts. The main span of the Asylum Bridge has inverted endposts in addition to the vertical "towers".

5. Camelback vs. Parker. I have never quite understood the logic for subdividing Parkers solely on the basis of nuber of panels.

Let's Play
Posted January 28, 2010, by Anonymous

>However, when you break them down into pieces they all revert back to one basic shape......the triangle.

Apart from the Vierendeel truss, of course.

I think this whole discussions just begs the question as to why you are categorising the bridges at all, other than database convenience. To a structural engineer, a key point is that the bridge itself does not "know" what type of truss it is, it's just a series of elements some in tension, some in compression, some of which alternate between both (that will particularly be true for the trussed suspension bridge examples), and some of which carry no real load at all (e.g. members introduced primarily to prevent buckling of other members). There are greater and lesser efficiencies for each type, but that's a design or sales issue.

The category is only relevant insofar as it identifies rarity (and hence contributes to historical significance - one bizarre hybrid has more value than a dozen Pratt or Warren trusses), or lineage (e.g. designs actually derived from Warren's patent rather than just those that happen to look the same but were arrived at by a different route).

For rarity, it would be best to split rather than lump, so that distinctiveness is given proper recognition, rather than anyone thinking "oh that's just another Pratt truss, we have plenty of them". For lineage, I suspect most of the bridges you've pictured here were not directly derived from the Pratt, Warren or other named concept, but independently arrived at. However, some may represent modifications introduced solely to avoid patent licensing, and if so, that would clearly be of historical interest.

Let's Play
Posted January 27, 2010, by Nathan Holth

Alright, I will comment on some but not all of these issues. I tried to keep my comments focused on facts, but with some (like the Queenpost) I could not resist laying out what is an opinion on my part.

Queenpost vs. Pratt - This question had driven me insane for years. I don't have a factual answer. However, I have always called them Queenposts, even with diagonals, because trusses of this configuration even with diagonals are extremely rare. I also had agreement with calling them Queenposts from Jim Stewart, a well-known historic bridge expert in New York State. One interesting thing to look for is whether the verticals are tension members or compression members. The tension members are more rare:

4-panel trusses - Yes, the bridge shown here can be called either a Pratt or a Warren. Unless its a state standard truss and you know of a larger example and what truss configuration the larger one follows, you might as well pick whatever you want. I list bridges like this as a Pratt, but thats just my choice.

5-panel trusses - I call them camelbacks whether through or pony. HAER does not mention in its truss diagram that the bridge must be a through truss.

Baltimore vs. Warren trusses - If the diagonals follow the Warren design, it is NOT a Baltimore. A Baltimore must follow the Pratt design. Sometimes however the term "Baltimore" is used to help people visualize what this "Modified Warren" truss design looks like. That may be why the Arkansas inventory listed this as a Baltimore. But technically, they were incorrect.

Subdivided Warren trusses - Doesn't matter how many verticals they are, these are all called subdivided. Bridges like this are double-intersection Warren truss bridges. I have seen the mis-labeled as lattice trusses. The term "lattice" truss specifically refers to those Warren truss bridges that go beyond the double-intersection (like a Quadrangular). Yes, the truss is functional. They were lightweight "budget" truss bridges marketed to townships as an economical alternative to the more traditional truss bridges. It is possible that bridge companies marketed them as "lattice trusses" which may explain where this description came from, but I strongly advise against calling them Lattice trusses today because most experts will think you are talking about bridges like Warren Quadrangular truss bridges.

English Center Bridge - What kind of bridge is this you ask? Its the English Center Bridge. Need I say more? :) Its obviously an experimental bridge type, which is why it defies classification. I would call it a suspension/truss hybrid. It will be listed as a suspension when I add it to

Victorian Footbridge: If you are not thinking of this as a suspension bridge you need to think of it as a cantilever bridge. Compare this bridge to a cantilever on the Delaware River: I cannot speak to the odd configuration of the suspended span in cantilevers like the Asylum Bridge, although its obviously designed to create a uniformity in the top chord. The term "Reverse Parker" is descriptive of the form, although I am unsure how widespread the term use is.

Let's Play
Posted January 27, 2010, by Anthony Dillon (spansaver [at] hotmail [dot] com)

This is the real beauty of truss bridges. They can be: inverted; reversed; variant-ed; double-intersected; triple-intersected; quadruple-intersected; sub-divided; half-hipped; countered; hybrid-ized; braced..........and on and on.

However, when you break them down into pieces they all revert back to one basic shape......the triangle.