Chatterjee and Templin 2004, review, part II (long)

[david peters <davidrpeters@earthlink.net>]

Posture, locomotion, and paleoecology of pterosaurs
Geological Society of America
Special Paper 376, 2004   64pp.
Chatterjee, S. and Templin, R.J.

In part II of this long review we start with a heading entitled: Wing Design, 
Comparative Morphology of the Wing

1. Chatterjee & Templin (C&T) compare the wings of bats, birds and pterosaurs. 
Basic stuff. C&T state: â??Both pterosaurs and birds belong to the clade 
Archosauria...â?? This statement comes four years after Peters 2000 showed that 
pterosaurs do not belong in the clade Archosauria. No argument is given. More 
importantly, no support is offered. Dropping pterosaurs into the Archosauria 
has always been by default and a bad fit.

2. C&T follow that statement with: â??As a result, pterosaurs and birds show 
closer phylogenetic similarity, as reflected in their anatomy and lifestyle.â?? 
What similarity??? Their closest common ancestor, assuming an archosaurian 
origin would have been erythrousuchian, according to the lastest from the 
â??Pterosaurs Are Archosaurs Peopleâ??. A diapsid opening and an antorbital 
fenestra is no guarantee of a similar anatomy and lifestyle. Strangely, almost 
every _other_ analogy in the book indicates that C&T think pterosaurs and bats 
have a more similar anatomy. Hereâ??s a quote, â??Here we prefer the bat model 
for pterosaur wings...â??

3. C&T follow Unwin et al. (1996) in supposing that the pteroid is a modified 
first distal carpal. Again, a bad guess arrived at by default. No evidence is 
given. Peters (2001) showed that ithe pteroid is derived from a central carpal. 
The first distal carpal remains lined up behnd digit I. The central carpals are 
missing having migrated anteriorly (medially) in nonvolant precursors.

4. Figure 12 shows the standard wing planforms of early and later pterosaurs 
based on Sordes and Pterodactylus.  Again arguments and drawings made by Peters 
(2001) against these very same specimens were not argued against. Simply 
ignored. (and yes, C&T reference the work.) 

5. C&T report, â??In bats, the large membrane in front of the wing can be used 
as a flap, another adaptation to low-speed flight.â?? Flaps always go on the 
trailing edge of the wing. Those on the leading edge are called slats.

6. C&T state, â??No [pterosaur] specimen shows a clearly defined trailing edge, 
attached unambiguously to the sidewall of the body.â?? Peters (2001) showed a 
number of examples of just such a wing, and Sordes is another good example of 
this. No argument was given by C&T. More samples have come to light since then. 
Not one is different.

7. C&T follow this with: â??[in the narrow chord wing]...the long trailing edge 
would remain free and dangling during flight without any mechanism of tensing 
the membrane.â?? Peters (2001) and Schaller (1985) showed that the wing is 
tensed between the wing tip and the elbow. No argument was given by C&T. 

8. C&T report: â??Without leg attachment, there would be no angle of incidence 
control and alteration of the wing planform.â?? First of all, most of the long 
wing is beyong the control of the hind leg, even if attached. Second, the angle 
of incidence control would be handled by shoulder muscles for the most part, 
but also by the uropatagium beween the legs and tail acting in flight as a 
horizontal stabilizer in an airplane, pitching the whole body.

9. C&T report: â??...the narrow, â??bird-likeâ?? wing without any leg 
attachment would make the control of ptich and camber more difficult.â?? First 
of all, birds donâ??t have any trouble controling pitch and camber with their 
narrow â??bird-likeâ?? wings. Second, a toy ornithopter has no trouble flying 
with a widely varying pitch and camber. Camber in a pterosaur wing could be 
controlled by the elbow which rises above the plane of the wing and curves the 
membrane like a circus tent, as in bats, but ornithopter toys show that even 
this is not necessary. 

10. C&T continue, â??...it would decrease the wing area slightly, thus 
increasing the wing loading and bending stress on the spar.â?? C&T are grasping 
at straws here. As a matter of fact, a narrow wing has much less drag (think of 
a glider), and whatever the wing loading was, pterosaurs handled it.

11. C&T repeat Unwin and Bakhurina (1994) in stating: â??[Pedal digit V] 
supported, tensed, and manipulated the uropatagium analogous with the calcar of 
bats.â?? No one, whether in 1994, nor in 2004, has shown how this occurs. What 
stretches, what rebounds, what are the limits, what are the abilities? This one 
Nature paper has bamboozled more scientists into accepting its statements 
without evidence and without testing them. A shame. Tischlinger & Frey (2002) 
report on a perfectly preserved uropatagium extending from the distal metarsal 
V to the tail base. Thatâ??s the pattern in all pterosaurs. This also was not 
reported or argued by C&T, but the paper was referenced.

12. C&T report that â??...the wing of a pterosaur... is curved to produce 
lift.â?? I donâ??t think camber has ever been preserved in stone. This might be 
wishful thinking. 

13. Figure 15 copies Bennett (2000) and his deep chord wing model bearing no 
resemblance to the reality of the Zittel wing that shares the figure. According 
to Bennett and now according to C&T: The propatagium attaches to the neck( 
which is false). The trailing edge of the wing is oriented toward the hips 
(which is false). And the elbow is over-extended. It is paramount that 
paleontologists follow Nature (the maternal figure, not the magazine) and not 
make up extended wing membranes and such just to please some horrible bat-wing 
paradigm (which Padian warned about over ten years ago and never fully heeded). 
By the way, Peters (2001) argued against these exact drawings with figures that 
were more faithful to Nature and solved more problems. 

14. C&T report that, â??In water spiders, small hairs on the body surface trap 
air and act as a water repellant. It is likely that pterosaur hair might have 
functioned in similar way [sic].â?? I think there are issues of scale here 
regarding water tension and beading, arenâ??t there? If the hairs are 
hydrophobic they will not get wet. 

Wing adaptations for powered flight

15. C&T report, â??Unlike bats and birds, the clavicles are absent in 
pterosaurs.â?? Wild (1993) identified pterosaur clavicles on the leading edge 
of the sternal complex.

16. C&T report, â??The scapula and coracoid... are elongated, L-shaped and 
fused together.â?? They are not fused together in a majority of pterosaurs. 
Fusion is restricted to certain clades so it is a phylogenetic character. It 
has not yet been shown to be an ontogenetic character despite Bennettâ??s 
pronouncements.

17. C&T report, â??In Cretaceous pterodactyloids, the upper end of the scapula 
fits into a socket of the notarium to make the joint immobile.â?? Wrong on two 
accounts. Not all Cretaceous pterodactyloids had a notarium. Think of 
ctenochasmatids and many of the smaller germanodactylids. The notarial joint 
was a ball-and-socket joint, with limited mobility, but still some.

18. Continuing: â??The coracoid is slightly longer than the scapula and is 
braced strongly against the transverse sulcus of the sternum.â?? Here C&T are 
thinking of their favorite pterodactyloids, not all of them. Some (tapejarids, 
dsungaripterids, some germanodactylids) have a short coracoid.

19. Continuing: â??The sternal plate is wide and has a shallow ventral keel.â?? 
But not in Istiodactylus (deep keel). And not in Zhejiangopterus (narrow 
plate), among others.

20. Continuing: â??In pterosaurs, there is a prominent acrocoracoid process on 
the coracoid to serve the same pulley function of the supracoracoideus muscle 
(Padian 1983). Yes, in Anhaguera and its kin there is a similar structure. No 
in most other pterosaurs, suggesting that this is not a case of basic 
convergence. Itâ??s the valley thatâ??s important for determining whether a 
pulley function is present or not â?? the valley between the process and the 
rest of the scapula. There is no valley in Dimorphodon, according to Padianâ??s 
drawing. Just a slight convex rise. Again wishful thinking in 1983. Not tested 
in 2004. Bennett also rebuilt pterosaur musculature without a pulley-like 
supracoracoideus in 2003 and he used Anhanguera as an example.

21. C&T report: â??â?¦by flexing the knuckle joint during the upstroke to 
reduce the drag.â?? Unfortunately, according to their wing model, this would 
relieve all the tension in the wing, turning it into a high-drag flapping flag. 
A sharp posterior bending of the wing works with birds. It might not work with 
pterosaurs. Does it work with bats? Not sure. In any case, the narrow, 
glider-like wing of pterosaurs already has a high-aspect ratio, which results 
in a high lift/ low drag ratio. Flexing the knuckle as much as is shown in 
Figure 17 would actually do more to destroy lift, rather than drag, which would 
be useful in a dive.

22. C&T report that the base of the wing could not close any closer than 35º 
in Anhanguera. Thatâ??s unfortunate, because in Istiodactylus, Pteranodon and 
any number of smaller pterosaurs the wing finger is sometimes found closed 
completely. Otherwise any wind on a grounded pterosaur is going to billow those 
membranes.

23. C&T hypothesize: The propatagium and pteroid bone might have acted as an 
elastic strap to keep the body close to the trunk [of a tree].â?? Peters (2001) 
referenced Brown, Baumel and Klemm (1995) who found that the propatagium in 
birds passively prevents hyperextension of the elbow during flight as the wing 
tip is subject to drag. The propatagium develops in leaping arboreal mammals 
and it appears in Longisquama, another arboreal leaper and a sister taxon to 
pterosaurs. Here the membrane forms a parachute to slow descent. Arm muscles 
are used to keep the body close to the tree, not elastic straps.

24. Figure 19 , the flight muscles of Anhanguera, is a direct copy from 
Wellnhofer 1991. Notably absent in both renderings are muscles that might 
connect the anterior dorsal spines to the scapula. There should be layers of 
them.

Aerodynamic Constraints

25. Figure 20 shows a number of pterosaurs in various views and scales in 
silhouette in order to compare wingspans and masses. Many were traced from 
various popular and childrenâ??s books. Why didnâ??t C&T show a ventral view of 
each of their wings to show they actually did the work? 

Much of the middle part of this paper goes into the mathematics of flight, 
Templinâ??s specialty. I note that because C&T chose a deep chord wing model, 
as every study before them did, their regression lines and scatter clouds were 
all a bit off from the bat and bird lines and clouds. A shallower chord would 
have brought all the data closer together. 

26. C&T state: â??In a shallow glide, in still air, there is no propulsive 
thrust, but there is a component of gravity acting forward along the sloping 
flight path.â?? In flight training we learned that gravity always acts 
downward. Inertia or momentum carries the object forward until friction, or the 
surface of the earth, stops it. 

27. Figure 21 has the standard wing shapes missing from Figure 20 presented in 
ventral view, along with a glider shape, a hangglider and a spread-eagle human 
for scale. Notably no pterosaurs have uropatagia here. Why not? In addition, 
the round tip wing shape C&T espouse earlier is  absent in all these pictures 
(only the largest preserve enough detail to tell). Why didnâ??t they fix this? 
It only promotes an image they know to be false.

28. Figures 22 and 25, once again, if C&T would reduce the wing chord all the 
pterosaurs would line up better with all the birds.

29. Table 5. Eudimorphodon ranzii, a specimen that has no wings, is species 
number one in a basic terrestrial takeoff data table. How can this be? Maybe 
theyâ??re thinking of another Eudimorphodon, probably MPUM 6009.

30. Figure 26 shows the glide polars for various pterosaurs. Dorygnathus comes 
out as a better glider than Rhamphorhynchus despite the fact that it has a much 
shorter relative wingspread and a much smaller sternal complex. Curious.

C&T spend many pages on hovering, steady flying and formation flying. Difficult 
to comment on these mathematical hypotheses when the wing shape is wrong and 
aerodynamics are not my specialty. 

Takeoff and Landing

The question posed by C&T is simple: running start? or flying start?

31. C&T state: â??the heavier a pterosaur was, the more trouble it had during 
takeoff and landing.â?? Define trouble. Does that mean effort or power 
expended? Does that mean awkwardness? Obstacle avoidance? All of the above? Do 
large birds have trouble taking off or landing, or are C&T mentally picturing 
gooney birds in particular? Lots of storks, pelicans and eagles have no trouble 
at all it seems to me.

32. C&T state that large pterosaurs needed a headwind or sloping surface for 
extra power in addition to a takeoff run. Hmmm. I see an ideal ploy as a 
predator. Not downwind, but down hill from my prey. Hey, give the pterosaurs 
some credit! 

33. C&T state: â??Because of their long wings that were rooted to their hind 
limb, pterosaurs could not flap during a takeoff.â?? I have never lived in the 
Mesozoic, but that statement doesnâ?? t sound right. A creature with wings 
unable to flap them? When C&T wrote this statement they should have pushed 
their chairs away from their computers and gone back to the chalk board. Or 
better yet, go rent â??Winged Migration.â??

34. C&T state: â??Long and wide runways would be needed...â?? The recent 
abstracts from this Novemberâ??s SVP include a report of tracks documenting the 
landing of a pterosaur. Took one tiny hop and stopped. With such large wings 
and small bodies it is easy to imagine one half of one flap (from dorsal to 
lateral) would send a body flying. Thatâ??s pushing a lot of air straight down. 
Plus donâ??t forget those leaping hind limbs! 

35. Figure 33 shows the take-off strategy of Quetzalcoatlus using a headwind 
and a downslope (poor guy!). First of all, this Quetzalcoatlus doesnâ??t match 
the skeleton seen earlier in figure 4. The neck is too short, the beak is too 
long, the wings are too long, the legs are too short, the metacarpus is too 
short. Itâ??s horrible! Plus, the way itâ??s running, carrying its wings half 
open creates twin high-drag parachutes. No wonder it had to plow downhill. 

36. C&T note that large pterosaurs might have used ground effect. Thatâ??s 
good. They might have. But probably unnecessary. It would be good to re-ask 
this question when the wing shape is right.

Sexual Dimorphism and Aerodynamic Function of the Head

37. C&T state: â??Among flying vertebrates, the most unusual feature of 
pterosaur anatomy is its immense skull, often dominated by a huge sagittal 
crest.â?? Here C&T are openly displaying their bias again. Very few pterosaurs 
had an immense skull. Fewer still had a huge crest.

Sexual Dimorphism

38. Here C&T repeat Bennettâ??s (1987, 2003) subjective guess to allocate large 
crests to supposed males because they supposedly also had narrow pelvic 
openings. The one Pteranodon that has a crest and a pelvic opening preserved in 
one specimen (the Triebold specimen) has a short crest and a small opening. So 
no cigar. The pelvic opening samples provided by Bennett, when scaled the same, 
are about the same diameter. The large pelvis has a narrow opening. The small 
pelvis a larger opening, both about 4 cm. No advantage in passing eggs 
therefore. Turns out other characters of the pelvis turn this into a 
phylogenetic problem. The two sample pelves were not the same species or genus. 
He was comparing apples to oranges, or Pteranodon to Nyctosaurus. Again, C&T 
could have tested Bennettâ??s assumptions, as I did, and in half an hour 
discovered the fatal flaws. No, they accepted without testing. Large crests, it 
turns out to no oneâ??s surprise, are only found at the end of phylogenetic
 lineages of â??no crestsâ?? leading to â??small crestsâ?? leading to â??large 
crests.â?? 

39. Continuing: â??The crest was developed last in ontogeny during sexual 
maturity.â?? Kind of like Bambi? No. Females and juvies probably had them too. 
There is no way to tell gender in a pterosaur unless thereâ??s a baby 
alongside. Crest development late in phylogeny? Yes. And that can be 
demonstrated in a cladogram. Thatâ??s the best answer we have for now that can 
be scientifically demonstrated.

40. C&T state: â??...all pterodactyloids had a proportionately enormous skull 
that was longer than the body length.â?? The micro-pterodactyloids close to 
Scaphognathus did not have a skull > torso. In G. cristatus + dsungaripterids 
it was subequal to the torso, not longer, and certainly not longer than the 
post-cranial body length. So that means not â??all.â?? 

Aerial turns

41. C&T state: â??The primary function of the enormous head of pterodactyloids 
appears to be associated with a yawing or steering mechanism.â?? Birds donâ??t 
do this. Bats donâ??t do this. Airplanes donâ??t do this. Pterosaurs with small 
heads donâ??t do this. Many pterodactyloids do not have an enormous head. C&T 
are blindly repeating bad dogma and focusing on their personal favorites  
again. All flying things bank to turn, which tilts the lift vector to that side 
and the turn continues infinitely until an opposite bank is applied. 

Ecology and Evolution

42. C&T state: â??The origin of and phylogeny of pterosaurs are still poorly 
understood.â?? What can I say? They read Peters (2001) but they could not 
overcome their archosaur paradigm and refused to accept. Or better yet, test!

43. C&T hypthesize: â??During swimming they could raise their wings vertically 
upward like a sailboat using wind energy to travel large territory for 
foraging.â?? Their own drawings actually limit the ability to lift the humerus 
to 20º from vertical. C&T hypothesize a locking mechanism (never found) for 
soaring, but this isnâ??t that. This is different. So theyâ??re asking the 
floating pterosaur to raise its wing(s) to its most dorsal position and hold it 
stiff while sailing? Once again, show me. Seems implausible at first glance. 
Seems tiring too.

44. Hereâ??s a problem in logic  that could have been solved just by re-reading 
the manuscript: C&T state: â??The skeletal design of pterosaurs does not 
suggest a diving adaptation (Brower 1983). Similar to brown pelicans, many 
small and medium-sized pterosaurs (M < 8kg) with good flying ability could 
forage by _plunge diving_ from the air into the water.â?? [emphasis mine]

44. C&T report: â??Apparently, pterosaurs spent a great deal of time eating. 
The enormous skull and narrow snout facilitated the large intake of food.â?? 
Again, not all pterosaurs had an enourmous skull and narrow snout. C&T are 
picking personal favorites again. And like the pelican, some pterosaurs beak 
can hold more than its belly-can. So, maybe more time searching for food and 
enjoying the afternoon. Maybe not â??a great deal of timeâ?? actually eating. 
Not even sure that whales spend a great deal of time eating. Crinoids do. Boas 
do.

45. C&T report on the various hypotheses concerning the origins of pterosaurs. 
Scleromochlus is considered then rejected as too close to theropods (not true - 
crocs). Atannassovâ??s dissertation on two Dockum archosaurs is mentioned, 
again both are apparently crocodilian. Bennettâ??s 1996 â??radicalâ?? 
hypothesis that pterosaurs come out close to erythrosuchids is mentioned.  
Closer, but still no cigar. C&T mention my report on Sharovipteryx as a 
possible ancestor, and ignore Unwinâ??s work along the same lines. They also 
ignore the other three protorosaur taxa that would work if for any reason 
Sharovipteryx was made ineligible to wear the crown. Wildâ??s (1983) imagined 
parachute lizard is discussed by C&T as if it was a possibility. That model is 
wrong. Peters (2000, 2001, 2004) showed that pterosaurs descended from 
protorosaur bipeds and that the wings came last after almost all the other 
modifications were in place,  almost as in birds.

Body size

46. C&T report that pterosaurs range down to sparrow-size. Actually the 
smallest hummingbird or bat is the lower size limit.

47.C&T discuss K selection vs. r selection, then state: â??Early pterosaurs 
with small size showed preference for r selectin, where production of large 
numbers of offspring was insurance against juvenile mortality or environmental 
catastrophe. Later pterosaurs exhibited K selection, with large size and long 
generation of time where the full resources of environments were exploited 
safely.â?? This is nothing but guesswork told as fact. No broods or clutches 
have been found for pterosaurs. Furthermore, this was at the printer before the 
discovery of the single pterosaur in the egg which turned out not to be an 
embryo after all.

48. C&T discuss Copeâ??s Rule, reject Copeâ??s Rule then show the effects of 
Copeâ??s Rule. To do this, C&T plot a mere ten pterosaurs on a guesstimated 
weight vs. guessitmated phylogeny  graph to show Gouldâ??s (1996) 
â??Drunkardâ??s Walkâ??. To make the chart work terminal taxa were flipped left 
to right and some closely related taxa, such as Nyctosaurus and Pteranodon were 
separated. Not good science from the start. Thereâ??s no discussion of size 
squeezes, which have not yet appeared in the literature, except in this chart.

49. C&T discuss increased oyxgen levels in the Late Cretaceous leading to 
gigantism in pterosaurs. C&T â??believe that an increased oxygen supply may  
enhance repiratory efficiency... enabling them to takeoff without switching to 
anaerobic physiology.â?? As if thatâ??s a bad thing! 

50. C&T conclude by discussing various Late Cretaceous extinction scenarios, 
suggesting that â??huge tsunamis produced by the impact would have destroyed 
shallow-marine habitats across the globe, devastating the sanctuary for the 
Maastrichtian pterodactyloids.â?? While birds had ecological superiority over 
pterosaurs. This is more guesswork, estimating where pterosaur nests were 
stationed and what destroyed them all. 

51. C&T report: â??Large-bodied pterosaurs had two disadvantages: they 
generally had smaller populations and lower reproductive rates than 
smaller-bodied bird species.â?? Again, where does this data come from? This is 
guesswork. Not all pterosaurs were giants at the end of the Cretaceous. Donâ??t 
forget the nyctosaurs. 

In the end, Chatterjee and Templin have given us a largely uncritical review of 
past literature, some novel flight theory and little else that isnâ??t tainted 
with bad data and guesswork. I cannot recommend this literature. There is 
little here that need be referenced in future work. Largely disappointing. 

Padian and Bennett were among the colleagues who reviewed the work prior to 
publication.