Flightless pterosaur – svp abstracts 2013

Now it’s in the published literature… the first known flightless pterosaur

Sos 2428. The flightless pterosaur.

Figure 2. Sos 2428. The flightless pterosaur. Click for more information and in situ photos.

Peters (2013) wrote: “Jura Museum Solnhofen Sammlung (SoS) 2428 is a largely complete, crushed, Solnhofen pterosaur. It was previously considered another specimen of Ardeadactylus (formerly Pterodactylus) longicollum, neotype: Staatliches Museum für Naturkunde, Stuttgart (SMNS) 56603. However, a closer look reveals important differences. The skull is longer than the cervical series in SoS 2428, but not in Ardeadactylus. The slender cervical ribs are each a centrum length in SoS 2428, but they are much shorter in Ardeadactylus. The parasagittally compressed dorsal vertebrae comprise only 40% of the torso length in SoS 2428, but 66% in the more typical pterosaur, Ardeadactylus. Conversely, in SoS 2428 the robust sacral series extends for 60% of the torso, 34% in Ardeadactylus. In SoS 2428 the dorsal ribs, sternal ribs and gastralia are relatively twice the lengths of those found in Ardeadactylus. The pectoral girdle is gracile in SoS 2428, with a scapula and a coracoid half the width of those same elements in Ardeadactylus. The forelimb (wing) elements are likewise less than half the length and width of those in Ardeadactylus. The wing finger (manual digit 4) is further reduced relative to the rest of the wing. When folded, the unreduced first wing phalanx extends back to the carpus. However, the second wing phalanx is half that length. The third phalanx is half the second and the fourth is less than half the third. Thus, when folded, the distal tip of the reduced wing finger extends just to the elbow. By comparison, in Ardeadactylus the elbow meets the middle of the second wing phalanx and the two distal phalanges nearly double that length. In SoS 2428, the free fingers, digits 1-3, are not reduced. Matching the elongated sacrum in SoS 2428, the hyperelongated ilium extends for 60% of the torso length. However, the much smaller pubis, prepubis, ischium and femur are similar in size to those same elements in Ardeadactylus. In SoS 2428 the distal tibia and pes are not preserved. When reconstructed, SoS 2428 has a relatively longer and wider torso than any other known pterosaur. It also has a reduced wing, half the length and half the chord
of the wing of Ardeadactylus when scaled to the same torso length. Such a reduced wing and enlarged torso make the prospect of flight rather doubtful by comparison. Moreover, with such morphological differences, SoS 2428 is clearly a distinct genus.”

Earlier we discussed this specimen. The poster for the flightless pterosaur included all the detail possible including color photos in high resolution. Tracy Ford was kind enough to put up this poster because I could not attend the SVP symposium. He said a few people walked by saying they did not believe it.

Doubt? Not convinced?
Belief, remember, is in the realm of religion. In Science you can confirm or refute a claim by repeating the experiment or observation. In this case, if you “don’t believe” the traits presented in this abstract, I encourage you to go visit the specimen and apply whatever technique works best at pulling data out. Whatever you get, send it to me. Then we can discuss this together, whether refuting, modifying or confirming the claim here.

Belief, in science, should never part of the equation. If you have evidence refuting this claim, please bring it to my attention. Changes will be made. Otherwise, join the celebration. Finally, we have a flightless pterosaur!

References
Peters, D 2013. A flightless pterosaur. Journal of Vertebrae Paleontology abstracts 2013.

Advertisements

Flipped wingtips – Rio Ptero Symposium

Continuing our look at the Rio Ptero Symposium abstracts…

Hone et al. (2013) proposed a reconstruction of the tiny Rhamphorhynchus specimen, Bellubrunnus with wing phalanx four (m4.4) concave anteriorly — in other words, different from all other pterosaurs (in which m4.4 is either straight or convex anteriorly). Hone et al. proposed that odd reconstruction because that is exactly how the wingtip appears in the fossil (Fig. 1).

In an attempt at explaining away the autapomorphic concave anteriorly curvature, which would have played havoc with the wing membrane, Hone et al. (2013) promoted the idea that the wingtip might straighten out under load, pulled and straightened posteriorly by a stretching wing membrane under tension, the way an archer’s bow bends when the archer pulls back the bow string.

There’s some confusion here,
so let’s take the problem apart with DGS.

Figure 1. Bellubrunnus (center) in situ. The Zittel wing (left) and the WDC specimen (right) showing how the interphalangeal joints are typically pointed posteriorly, toward the membrane. The Bellubrunnus specimen appears to break that model, but note the wings have been torsioned so the dorsal side is shown in green and the ventral side appears in red. Confusing, yes!

Figure 1. Click to enlarge. Bellubrunnus (center) in situ. The Zittel wing (left) and the WDC specimen (right) showing how the interphalangeal joints are typically pointed posteriorly, toward the membrane. The Bellubrunnus specimen appears to break that model (see insets at bottom), but note the right wing has been flipped (ventral surface in red) and the distal half of the left wing has been flipped (ventral surface in red) based on the shape of the wing joints. This makes the wing tips typically convex anteriorly when flipped back to their in vivo orientation. 

Clue #1. The wing tips are not exposed in dorsal view
While the fossil is largely preserved in dorsal view, the left distal elements, m4.3 and m4.4 (in red) have been twisted axially 180 degrees, exposing their ventral surfaces. The right wing is exposed ventrally (in red) in its entirety. We know this because of the shapes of the joints. Examples (Fig. 1) from the Zittel wing and the WDC specimen illustrate what the unrotated wing joints look like. The largest, most pointed parts of the wing joints generally point toward the wing membrane. That’s basic pterosaur morphology. That’s how you can tell dorsal from ventral… most of the time.

The confusion arises
from the proximal process of the left m4.4 which is larger and more pointed — the opposite of most pterosaur wing joints. This is what Hone et al. keyed on. The right wing doesn’t show any expansion of the joint in either direction. So it is oddly uninformative. Basically Hone et al. chose that joint expansion clue on the left m4.4 to decide that joint was on the wing phalanx posterior rim, giving it a concave anterior bow. There’s a certain logic in that that I can appreciate. But then, we have to deal with the extreme oddity of concave wing tips!

Let’s look at some other related wing tips.

Figure 2. M4.4 from another Rhamphorhynchus, showing the joint can expand both anteriorly and posteriorly.

Figure 2. M4.4 from another Rhamphorhynchus, showing the joint can expand both anteriorly and posteriorly.

Other Rhamphorhynchus specimens (Fig. 2) demonstrate that the proximal joint of m4.4 can expand in both directions. This may solve the problem. A reconstruction (Fig. 3) also helps.

Figure 1. A reconstruction of the new pterosaur, Bellubrunnus.

Figure 3. A reconstruction of the new pterosaur, Bellubrunnus with wingtips (m4.4) convex anteriorly as in most other pterosaurs. Otherwise we’ll have problems with the trailing wing membrane.

A sister taxon – Qinglongopterus
The large pterosaur tree nests the small Qinglongopterus as a sister to Bellubrunnus and it also has a double-expanded proximal joint on m4.4 (Fig. 4). So, now, if you’re a paleontologist, you have to decide:

  1. did the preparator knock off the joint expansion on the concave side of the Bellubrunnus left wing during prep? … and also both joint expansions on the right wing?
  2. are the wing tips flipped ventral side up? (See Fig. 1).
  3. did Bellubrunnus have an odd sort of convex anterior wing with the joint expansion anteriorly and only on one wing?
  4. or did Bellubrunnus have a concave anteriorly phalanx, different than all other pterosaurs?

Hone et al. chose #4. Then they had to explain it away but employing the “under tension” hypothesis. Occam’s razor might have helped here. It’s like trying to explain away a folded wing membrane as “membrane shrinkage” during fossilization, which  Elgin, Hone and Frey (2011) did earlier.

Figure 4. Qinglongopterus, the sister to Bellubrunnus, has a  proximal m4.4 that is expanded in both directions. So, it happens to close relatives.

Autapomorphies
As I learned with regard to that funny little “anterior process” on the ilium of Cosesaurus that turned out to be the stem of the prepubis, autapomorphies can turn out to be mistaken interpretations. If you find an autapomorphy, a trait that no other sisters have, like a concave wingtip, it’s best to think through the possibility that you may have made a mistake. Unfortunately, Hone et al. did not explore all the possibilities, nor did they find the same trait on the other wing or other sister taxa. There is something odd about that one joint, but nothing odd about the convex anterior phalanx.

It’s always good to discover things, but potentially embarrassing when someone points out “kind sir, you have it upside down.”

Once pterosaur wings evolved, very little changed about them other than the relative sizes of wing bone elements. An anteriorly concave wing would have been quite a difference — and one that was reversed back again in descendant taxa, as no later Rhamphorhynchus specimens share this trait.

This also hearkens back to Bennett 2008
who claimed the pterosaur wing finger and membrane began as part of a supinated limb (in other words, not facing the substrate) — but lost the forward-pointing claw from the wing finger because it would have pointed forward hooking on nearby tree trunks — and the naturally concave palmar bow of the the terminal phalanx would have had to reverse to convex forward — along with turning the natural flexion of the metacarpophalangeal wing joint to hyper-hyper extension. All this completely imaginary hoo-hah is detailed here.

References
Bennett SC 2008. Morphological evolution of the forelimb of pterosaurs: myology and function. Pp. 127–141 in E Buffetaut and DWE Hone eds., Flugsaurier: pterosaur papers in honour of Peter Wellnhofer. Zitteliana, B28.
Elgin RA, Hone DWE and Frey E 2011. 
The extent of the pterosaur flight membrane. Acta Palaeontologica Polonica 56 (1), 2011: 99-111. doi: 10.4202/app.2009.0145
Hone DWE, Habib MB and Van Rooijen M 2013. Wingtips in Pterosaurs: functional and ecological implications. Rio Ptero Symposium 77-78.

.

 

Zittel Wing – Rio Ptero Symposium

Earlier we looked at another Rio Ptero Symposium abstract. There are several more to come.

Chris Bennett (2013) took a fresh look at the century-old Zittel wing of Rhamphorhynchus (Fig. 1). He determined that we need to flip our view of it. According to Bennett, those aren’t straw-like actinofibrils ribbing the wing that contract to make the wing extensible. Rather those are narrow spaces between popsicle-stick like actinofbrils that make the wing “inextensible.”

Figure 1. The Zittel wing simply flipped to give the impression of dips (upper image) or ridges (lower image) in the wing membrane due to shadows that the eye assumes to come from above the image.

Figure 1. The Zittel wing simply flipped to give the false image of dips (upper image) or true image of ridges (lower image) in the wing membrane due to light direction that the eye assumes to come from above the image. Note the round bones give away the true light direction. Inserts show Bennett’s new view (upper image) of wide aktinofibrils. Lower image shows traditional view of straw-like aktinofibrils that enable wing reduction.

Bennett notes, “actinofibrils were long flattened bands of keratin” and “The raised longitudinal strips are not the actinofibrils, but rather the grooves between the ridges are impressions of the actual actinofibrils and thus for 120 years we have had it upside down and backwards” and  “Thus its size and shape of the Zittel wing is essentially as in life.” Of course that flies against all the other examples of pterosaurs in which the wing folds to near invisibility.

This is not Bennett’s first foray into radical and irrational hypotheses. His autapomorphic interpretation of the flathead anurognathid skull and his completely imaginary hypothesis on pterosaur wing origins are among his earlier ventures. It’s hard to reconcile Bennett’s inextensible wing with the several other examples in the fossil record of extensible wings that collapse into almost nothingness.

On the other hand
in the same Rio Ptero volume Bennett successfully argued against the presence of an antorbital fossa and mandibular fenestra in basal pteros, as discussed earlier here.

References
Bennett SC 2013. If 6 was 9: Turning our interpretation of the Zittel wing upside down. Rio Ptero 2013 Short Communications. 22-25.

Alpine Swifts and Pterosaurs

Alpine Swift (Tachymarptis melba) on the wing for 7 months at a time! Check out that wing shape. Remind you of anything prehistoric?

Alpine Swift (Tachymarptis melba) on the wing for 7 months at a time! Check out that wing shape. Remind you of anything prehistoric?

An interesting NatGeo post on the Alpine Swift (Tachymarptis melba) and its incredible but true 7 months (200 days) on the wing (Liechti et al. 2013) raised my curiosity about what sort of wing must such a bird have?

Turns out to have a very pterosaurian wing (short chord version) in ventral view. Nat Geo reports, Their long wings make them fast and manoeuvrable, allowing them to scythe through the air in search of small insects and other “aerial plankton”.

And why do they fly continuously? Again, Liechti has speculations rather than answers. They may exploit food sources that other birds can’t touch, avoid predators by flying through the night, or stay beyond the reach of parasites like malarial mosquitoes. “These aren’t very convincing,” he admits, “but for sure, there’s a cost to staying in the air, so there must be a benefit.”

Swifts are apodids, famous for their tiny feet (they don’t perch that often). That takes us to the pterosaurs with tiny feet and spindly legs, the ornithocheirids (Fig. 2, 3).

Worth comparing for wing shape and foot size.
Evidently these large pterosaurs were likewise rarely grounded, based on their tiny feet and giant wings, especially compared to other pterosaurs.

Figure 2. The ornithocheirid pterosaur, Arthurdactylus. Note the tiny size of its feet.

Figure 2. The ornithocheirid pterosaur, Arthurdactylus. Note the tiny size of its feet and the huge wings. Like a swift, this pterosaur could have slept while on the wing. The spindly fingers were no good for grappling tree trunks.

Awkward on the ground.
Graceful in the air. This is the reason why the ornithocheirid humerus is so much larger than the femur – not the forelimb launch hypothesis! They put everything into their wings, which transport them to food. Their legs simply enable them to walk out of their eggs.

Figure 2. Arthurdactylus in dorsal view while flying. Note the knife-like wing shape,  that could be maneuvered, like that of a sail plane or swift.

Figure 2. Arthurdactylus in dorsal view while flying. Note the knife-like wing shape, that could be maneuvered, like that of a sail plane or swift. Wings back = less drag, greater speed.

Basically the pterosaur wing in ornithocheirds is a tapering cone, with a large diameter proximally and a tiny diameter distally. This has proved to be a very strong structure from outstretched traffic lights to fishing rods.

Outstretched to swept back
As in swifts, the wings of pterosaurs could have maneuvered in flight from strictly lateral to backswept. Each configuration has their own use, advantage and disadvantage.

References
Liechti, Witvliet, Weber & Bachler 2013. First evidence of a 200-day non-stop flight in a bird. Nature Communications.http://dx.doi.org/10.1038/ncomms3554

Pterosaur dipping and skimming – first for drinks, then for floating insects

Last week’s (Sept 18) Nova program on PBS entitled, “Earthflight” showed swallows, with their tiny little beaks, dipping for water while on the wing (see video here). Later the swallows dipped for floating mayflies. Others took quick baths by diving an inch below the water then reemerging without losing too much momentum.

From the BBC special on birds entitled, "Earthflight." Swallows drinking while on the wing.

Click to view. From the BBC special on birds entitled, “Earthflight.” Swallows drinking while on the wing. Note the teeny tiny beaks.

So why can’t certain pterosaurs, like Rhamphorhynchus and others, skim and dip (there is a spectrum between the two based on speed of attack) for surface-dwelling fish? Especially when a fish is found in the throat and belly? We looked at this earlier. The swallows, with their tiny wide, unspecialized for dipping beaks, demonstrate that a flying animal does not need a special beak shape.

Skimming pterosaur

Figure 2. Manipulating the bones of the fish-eating Rhamphorhynchus into a skimming configuration while staying airborne.

Humphries et al. (2007) tried to show that skimming would not be likely for pterosaurs. They used math. Sometimes math isn’t the key. They used morphological comparisons to the modern skimmer, Rynchops, which is ideally suited to high-speed skimming on windless ponds. Swallows demonstrate that’s THAT important. We considered pterosaur skimming earlier here.

Earlier we talked about the key: windspeed. It doesn’t equal groundspeed (or waterspeed) when flying into a breeze or steady wind. A steady breeze at the best glide speed can equal a hover over a particular spot. A little speed isn’t a bad thing either. I’m sure pterosaurs found the ideal circumstances and took advantage of them.

So, like birds, pterosaurs graduated from taking drinks on the wing, to taking floating insects on the wing, to taking surface-dwelling fish while on the wing and some even dived like gannets deep beneath the surface (gannets can descend 70 feet btw). That’s a nice variety of niches.

Nyctosaurus and Pteranodon
On this subject, it’s interesting to note, once again, that Nyctosaurus had a longer mandible and Pteranodon had a longer rostrum, both sharp like a sword and sharpened with a single sharp tooth in the tips of both jaws. Rhamphorhynchus (Fig. 2) does not have these anterior teeth, but has a rostrum and dentary tipped with keratin extensions.

References
Humphries S, Bonser RHC, Witton MP and Martill DM 2007.
 Did pterosaurs feed by skimming? Physical modelling and anatomical evaluation of an unusual feeding method. PLoS Biol 5(8): e204. doi:10.1371/journal.pbio.0050204 online

Pterosaur ornithopters: lessons learned

Following in the success of the Dr. Paul B. MacCready‘s 1985 flying Quetzalcoatlus ornithopter (Fig. 4), a few years ago there was an attempt at getting another very complex pterosaur ornithopter to fly.

Margot Garritsen is a Dutch engineer and Stanford professor who led a team intent on building a flying pterosaur based on Paul Sereno’s ornithocheirid from the Sahara. They were counting on greater success with lighter materials and a more accurate wing movement with not one, but five wing joints for flight control. Several paleontologists were team members and Hall Train provided some of the mechanics. So it had everything going for it. The project was featured in the IMAX film “Sky Monsters.”

Figure x. The Stanford pterosaur ornithopter moments after dropping from its mothership. On this second attempt all the fur and non-essential material had been removed.

Figure 1. The Stanford pterosaur ornithopter moments after dropping from its mothership. On this second attempt all the fur and non-essential material had been removed. A removable horizontal stabilizer with twin rudders is added as a sort of stabilizing tail. Note, this is a deep chord wing membrane configuration, which pterosaurs did not have.

Unfortunately
the new and improved ornithopter failed to flap and failed to fly.

Another inventor, Kazuhiko Kakuta
using a much simpler design (Figs. 2, 3), created a successfully working ptero-ornithopter.

Cheaper. Simpler. Less accurate.
Actually, almost nothing is more pterosaur-like than bird-like here other than the fashioned crest. The key here appears to be the successful creation of sufficient thrust and lift without a cambered airfoil — as in any toy bird-like ornithopter.

For those interested ornithopters are explained here.

Figure 1. Pterosaur ornithopter. This model flies well and for good reason.

Figure 2. Pterosaur ornithopter. This model flies well and for good reason.

An efficient flapping wing must be able to flex and/or rotate: if a static wing is kept at the same angle while moving up and down, it will produce no net lift or thrust. Flexible wings can attain efficiency while keeping the driving mechanism simple. In Ornithopters its the ventral and dorsal curling of the wing during flapping that changes the wing shape and creates lift and thrust.

Read about the model maker here with his other pterosaur YouTube videos listed.

Figure 2. Click to see video. This pterosaur ornithopter folllows the basic plan of bird ornithopters in having a stiff leading edge and a flexible trailing edge. There's no need for complex flapping cycle. Up and down works pretty well.

Figure 3. Click to see video. This pterosaur ornithopter folllows the basic plan of bird ornithopters in having a stiff leading edge and a flexible trailing edge. There’s no need for complex flapping cycle. Up and down works pretty well.

Most ornithopters have extremely simple motions and deep chord wing shapes.

What would happen if the wing had a camber, a narrow chord and a spoon-shaped wing tips, as in pterosaurs? So far, except for the MacCready invention (Fig. 4), no one has built a short chord. long wing ornithopter and even the MacCready invention did not have the proper pterosaur wing shape and leg configuration.

So there’s an opportunity here to do something great for an engineering student: replicate a real pterosaur and make it flap using simple ornithopter techniques.

Figure 3. Quetzalcoatlus model ornithopter by Paul Macready getting walked to its take-off point.

Figure 4. Quetzalcoatlus model ornithopter by Paul Macready getting walked to its take-off point. The tucked in legs are based on the bird-like hypotheses of Dr. Kevin Padian, now widely regarded as wrong. No fossils preserve this configuration. Rather the legs would have been more or less splayed in flight.

Dr. Paul B. MacCready is famous for creating a dang big ornithopter the size and shape of a Quetzalcoatlus back in 1985. Here it is on YouTube. Here is a pdf of the project. It flew very successfully. There’s a Popular Science article here about MacCready’s work.

Still…
It would have been better to extend those hind limbs like horizontal stabilizers on airplanes (Fig. 5), but they were listening to Kevin Padian back then and he saw pterosaurs as very bird-like. Now that we know they were more lizard-like, pterosaur configurations have changed. 

Rhamphorhynchus model by David Peters

Figure 5. Rhamphorhynchus model by yours truly. Note the narrow chord long wings and feet splayed like a horizontal stabilizer. The raised elbows produce more camber proximally. The tail is an unnecessary secondary sexual characteristic.

For a change of pace, here’s a video that shows a small simple pterosaur-shaped airplane powered by propellers. So basically, it’s an airplane.

Beipiaopterus and all that messy soft tissue – what is it??

Beipiaopterus (Lü 2002, 2003, Figs. 1, 2) is a roadkill pterosaur fossil that appears to have been physically crushed prior to burial. The radius and ulna are broken in half. There is a mass of soft wing tissue that definitely attaches to the hind limb and this has been used to “prove” that the wing membrane attached broadly to the tibia. That mass of soft tissue doesn’t really look like typical wing tissue, like the Zittel wing, unless it has been wadded up like an old receipt. There are also some long hairy parts, unlike typical wing membranes.

I’ve often wondered what was going on there with that soft tissue. Here we’ll take a good look at Beipiaopterus using DGS and see where it might fit on a reconstruction, something Lü (2002, 2003) did not attempt.

Figure 1. Beipiaopterus in situ plus the Lü 2003 interpretation of the elements. M4.1 is actually m4.1+2 based on comparisons with sister taxa and the wing has the requisite 4 wing phalanges + the ungual.

Figure 1. Beipiaopterus in situ plus the Lü 2003 interpretation of the elements. M4.1 is actually m4.1+2 based on comparisons with sister taxa. That means the wing has the requisite 4 wing phalanges + the ungual. Distal m4.1 is missing, here filled with plaster. The rough tracing by Lü misses many details that help to describe this taxon.

At first glance, it’s not at all clear what is happening with the soft tissue of Beipiaopterus, but manifestly clear that there is a mass of soft tissue there. So how can we interpret it?

Let’s make a reconstruction
If we put Beipiaopterus back together again (Fig. 2), perhaps we can find suitable places to move the soft tissue back into.

Figure 2. Beipiaopterus in dorsal and lateral views. In lateral view you can see the soft tissue mass is too large for the hind limb, but a uropatagium is present (in lavender), which is the part attached to the hind limb. The rest (in light blue) seems to come from the tip, middle and inboard sections of the wing. The tip has strong aktinofibrils. The middle less well defined. The proximal portion appears to be quite hairy. If anyone has a better idea, please send it in.

Figure 2. Beipiaopterus in dorsal and lateral views. A and B areas correspond to figure 1. In lateral view you can see the soft tissue mass is too large for the hind limb, but a uropatagium is present (in lavender), which is the part attached to the hind limb. The rest (in light blue) seems to come from the tip, middle and inboard sections of the wing. The tip (C) has strong aktinofibrils. The middle (B) less well defined, but attached to m4.3 in figure 1 as it is here. The proximal portion (A) appears to be quite hairy is located near the femur in situ. This could represent body hair. If anyone has a better idea, please send it in.

Given an understandable playing field (by straightening out the bones in a reconstruction), we can start placing the soft tissue “turf” where it seems to fit best. After all the soft tissue is a wadded up mess, just like the rest of the skeleton. Section A seems to be proximal hairy material. Section B seems to be midwing material and is attached to manual4.3. Section C seems to be wingtip material. Section D is definitely uropatagial as it attaches to the tibia and has a trailing edge.

Deep chord wing proof?
No. This specimen does not prove the wing membrane attached to the tibia, as Lü (2003) reported. Part of the wadded up membrane indeed does attach to the tibia, but that’s the uropatagium and it’s narrow. The rest is displaced wing tissue. We’ve seen displaced wing tissue on Sordes. And here’s an example of wing tissue just starting to tear from its mast (Fig. 3). Note how it starts to fold on itself.

Figure 3. A specimen of Rhamphorhynchus demonstrating how the wing membrane starts to separate from the wing finger during taphonomy.

Figure 3. A specimen of Rhamphorhynchus demonstrating how the wing membrane starts to separate from the wing finger during taphonomy.

References
Lü J-C. 2002. Soft tissue in an Early Cretaceous pterosaur from Liaoning Province, China. Memoir of the Fukui Prefectural Dinosaur Museum 1: 19-28.
Lü J-C 2003. A new pterosaur: Beipiaopterus chenianus, gen. et sp. nov. (Reptilia: Pterosauria) from Western Liaoning Province, China. Memoir of the Fukui Prefectural Dinosaur Museum 2: 153-160.
Lü J-C, Kobayashi Y, Yuan C, Ji S and Ji Q 2005. SEM Observation of the Wing Membrane of Beipiaopterus chenianus (Pterosauria). Acta Geologica Sinica 79:6 766-769.