[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

coelurosauravus revisited

Senter's recent attempt to lump Longisquama with Coelurosauravus got me 
interested in that Permian gliding diapsid, close to the base of all living 

The second most recent paper on Coelurosauravus was in 1997:

Frey, Sues and Monk 1997
Gliding Mechanism in the Late Permian Reptile Coelurosauravus
SCIENCE -VOL. 275 - 7 MARCH 1997

pdf available at: 

In this paper FS&M demonstrated that unlike other gliders and flyers, 
Coelurosauravus developed its wing via neomorphic dermal ossifications without 
modification to the ribs or limbs, as in other flyers. They reconstructed a 
wing in which the rods had a radial disposition originating near the humeral 
glenoid and posteriorly no further than 1/3 of the torso. The leading edge of 
the wing in their view formed an angle of 75 degrees with the sagittal plane 
and the posteriormost wing rods were oriented parallel to the parasagittal 
plane. The trailing edge of the wing in their reconstruction terminated well 
anterior to the hind limb. They also reported that there was no one-to-one 
relationship of the rods and ribs due to the much larger number of rods than 

I had no reason to doubt this interpretation until I started diving into the 
literature and tracing the specimens. Although FS&M got the main point right, 
(dermal ossifications as wing supports) some of the details may be arguable.

1. In my reconstruction (see www.pterosaurinfo.com/coelurosauravus.html) the 
wing is broader than FS&M reconstructed it, with a leading edge extending at 
right angles to the sagittal plane. This reconstruction was created by rotating 
each rib in Photoshop, rather than redrawing it freehand as FS&M seem to have 

2. If the posteriormost rods are left parallel to the sagittal plane, they 
extend past the posterior pelvis, which means the leg is under the wing.

3. Rather, if the posteriormost rods extend laterally, they can be 
reconstructed completely anterior to the hind limb, producing a slightly 
narrower wing with rods more closely spaced. Probably a better design for that 
reason alone -- but who knows?  Jim C.??

4. There are 23 rods in total per side. The first 11 emanate from the first 
dorsal rib and form a large bundle proximally, as HS&M noted. However, the 
posterior 12 rods appear to be associated with the next 12 dorsal ribs, 
one-to-one. Thus half are not radially oriented as HS&M reported in 1997. About 
half are though.

5. SMNK 2882 PAL might be a mother. When all of the bones af the main specimen 
are accounted for, more remain. In both specimens the skull is the only 
disarticulared element. I'd be interested to see if others can reidentify the 
bones, especially that unique skull, any other way. 

6. The affinities of Coelurosauravus are still an open question. The 
metacarpals and metatarsals decrease in length laterally from II to V. When we 
find another taxon with this pattern, we might have a clue.

7. The hands and feet, to no one's suprise, appear to be webbed.

8. The pelvis, always hard to see and never rendered previously, is found here 
to be quite similar to that of Petrolacosaurus with the addition of a bit of 
anterior process on the ilium. Due to this process, if facultatively bipedal in 
the manner of the Australian Frillneck, then the spreading of the rib rods may 
have had another use -- prior to its use as a gliding membrane -- to increase 
the apparent size of Coelurosauravus in the face of rivals and predators.

Interesting notes on reproduction in Draco volans:

It is not known exactly when reproduction occurs, but it is assumed to be in 
December and January. Males, and occasionally females, show several displays. 
These include the spreading of the wings and a bobbing motion of the entire 
body when the two are in close proximity to each other. The male will also 
spread his dewlap to a fully erect position and then circle the female three 
times before copulation. The female will only show display patterns to stop or 
prevent copulation (Hairston, 1957). The female Draco will build a nest for the 
eggs by forcing her head into the soil to create a small hole. She will then 
lay five eggs into the hole and cover them with dirt, packing the soil on top 
of them with a patting motion of her head. For approximately twenty four hours, 
the female will guard the eggs fiercely. After this period, no further guarding 
occurs. Incubation of the eggs take approximately 32 days (Card, 1994).

More later,

David Peters
St. Louis