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

Who says dromaeosaurs can't fly?

Got the Czerkas book and have been looking over the comments, particularly 
concerning Cryptovolans. 

It is true that the papers could have been better edited and had some 
analytical problems. But the data they contain is vital and ground breaking, 
and despite the flaws the analyses are superior to those commonly presented 
in more established venues.  

Steve's analysis that Crypto was a fully developed flier is entirely correct. 
Both arms are articulated. There are very large (far bigger than 
Caudipteryx), but badly preserved, primary feathers anchored on the type's 
right hand, with some smaller, well preserved feathers attached to the distal 
leading edge of digit II in the avian manner. From the left radius and ulna 
emerge a well developed array of middle wing feathers. The left hand lies 
immediately before the left thigh. The very large, well developed, well 
defined, and well preserved feathers immediately behind the thigh are a 
distal continuation of the left arm's wing feathers, and are in the exact posi
tion and have the exact form to be the fully articulated, partly folded 
primaries of the left hand. Pretty as can be. The feathers are not anchored 
to the left femur, because they are present both in front and behind it, 
because there are no thick anchor shafts immediately behind the femur, and 
because the feathers at the femur are clearly at their broad chord 

The more distal array of feathers associated with the left tib-fib are much 
less well defined, and are not in position to be primaries from the left 
hand. They do not extend in front of the tibia, do taper approaching the 
tibia, and there are anchor shafts immediately posterior to the tibia. These 
feathers are distal leg feathers. It just so happens that this is the same 
arrangment usually found in modern flying predatory avepods. 

Indeed, I have a photograph of a standing bird of prey that shows exactly 
this pattern, with very long (as long as Crypto's) lower leg feathers 
projecting posteriorly, and the folded wing primaries immediately above and 
projecting far behind the thigh. Kill that bird and flatten it on its side in 
the same position with a 57 Chevy, let the road kill rot some and you'd have 
some peering down and claiming the primaries are thigh feathers:-) The Berlin 
Archy also has well developed lower leg feathers, albeit not as long as 

The lower leg feathers are too poorly preserved to tell their degree of 
symmetry or non, and I do not know what their condition is in flying birds. 
In any case modern birds with long leg feathers do not use the legs as a true 
"forth wing", and there is no reason to think flying dromys did either. How 
birds use such feathers if they do would be interesting to know. 

Steve is entirely correct that Crypto's wing primaries are strongly 
asymmetrical. The feathers at the distal end of right finger II are extremely 
so, with the main shaft nearly at the leading edge. Don't obsess over quirks 
in the text, look at the excellent photos the Czerkas's went to all that 
trouble and expense to give us. Multiple examples of the large left primaries 
show that the trailing surface is twice as broad as the leading, it can 
easily be measured off those big gorgeous photos. That the feathers are so 
asymmetrical verifies that they are primaries rather than thigh feathers. 
Such asymmetry occurs only when feathers are adapted to generate thrust 
and/or lift, there are no exceptions. 

The big wing primaries were much more strongly mounted on the robust, 
flattened finger II of Crypto than in Archy, in which the same finger remains 
slender and only has a mere hint of the expansion needed to solidly anchor 
the flight feathers. Crypto also has a strongly bowed metacarpal III that 
helps support the primary fan like birds, but not Archy. 

I did a restoration of the Crypto skeleton and wing sufficient to estimate 
proportions, mass and wing area. Mass is about two and a half times that of 
the largest Archy. The leg is significantly longer relative to the body than 
in Archaeopteryx. The arm relatively a little shorter than that of 
Archaeopteryx, but as Steve notes the primaries are somewhat longer than the 
hand which is not surprising considering the well developed finger flanges 
for anchoring them. So total wing span and area are about the same, maybe a 
little above, that seen in Archy. Plot the mass and span on Append Fig 18G in 
DA and Crypto falls right on the main avian line, so it had plenty of wing to 
fly with. 
As for power, Crypto has a fully ossified sternum about a half dozen times 
larger relative to the body than that of Archy, and fully ossified sternal 
ribs to firmly anchor the sternum to the ribcage. Ossified uncinate processes 
may have further strengthened the ribcage, again Archy lacks them. 

So we have an unambiguous basal dromaeosaur that has a fully developed, lift 
generating wing as relatively big and well developed and aerodynamic as that 
of Archaeopteryx, but with the largest flight feathers more securely mounted 
than in the early bird. The flight power system of the dromy is clearly much 
better developed than that of the urvogel. That Cryptos arms were somewhat 
shorter than its exce[tionally long legs means nothing since the arms of 
another living predatory avepod, the secretary bird, are far shorter than the 
legs. The dromy's tail is at least as sophisticated, in a pterosaur manner, 
as that of Archaeopteryx. 

Everyone get used to it. At least some basal dromaeosaurs could fly baby, and 
better than Archaeopteryx. In no way was Crypto a protoflier, it was a fully 
developed flier whose performance approached and equalled that of pterosaurs 
and confusiusornithids that lack deep keeled sterna but have well developed 
wings. The dromaeosaur should have been able to take off from level ground, 
climb and fly a substantial distance, and land high in a try. It may have 
used flight as a regular attack mode. Imagine sickle-claws perched high in 
trees, waiting for some hapless prey to walk through the neighboring glade 
only to be pounced upon from above! Screaming dinosaurs from the air! 
(Research into stomach contents indicates that cladists were considered 
especially tasty). Since the Jehol is churning out fossils like candy out of 
a Hershey factory we will get more and more fossils that will only make this 
increasingly clear, so there is no point making a fuss about it. You don't 
want to be like those who went on and on about how dinos don't have feathers 
and those fibers are really just frayed collagon, now do you? If you do, the 
burden is upon you to demonstrate that Cryptovolans was not a competent 

Readjust your minds to the new reality. 

As Steve notes sinornithosaurs may have been big winged fliers. If anything 
the arm was longer relative to the body, and they have the robust, flattened 
central finger and bowed outer metacarpal suitable for supporting large 
primaries. The sternal plates were similarly very large, although not fused. 
My guess is that new sinornithosaur specimens will turn up with great big 
glorious wings.  Cannot say much about Microraptor since not much of the arms 
is preserved, we'll found out soon enough. The type Bambiraptor has a real 
big but unfused sternum, and very long arms albeit not as relatively long as 
cryptos and sinornithos. Bambi lacks the flattened central finger. May have 
had some flight capability, especially for leaping onto prey from trees. 
Looked at our buddy Deinonychus but the arms are just to short to support 
wings large enough to be really effective. Sternum is somewhat smaller and 
the central finger is not flattened too. Might have been enough wing to 
extend long leaps. Sure would like to know the size, symmetry and fraying of 
the primaries.    

I've noted that the pterosaur-like tails of dromaeosaurs are probably 
pterosaur-like (with hyperelongated distal prezygapophyses and chevrons) 
because the tails evolved in fliers. The tail form is, after all, not present 
in any unambiguous nonfliers. A prediction of this hypothesis is that there 
should be avepods more derived than Archaeopteryx with long, dromaeosaur-like 
tails. Crypto fits this bill to a tee. In Nature the description of the new 
basal bird Jeholornis states that its tail has "unexpected elongated 
prezygopophyses and chevrons, resembling that of dromaeosaurids." Unexpected 
only if one is locked into the conventional hypothesis. Yet again a 
prediction of the neoflightless hypothesis is fulfilled. Birds did experience 
a flight stage in which the tail functioned in the same manner as long tailed 
pterosaurs. It will be interesting to see the distal tail of sickle-clawed 

Since all known avepod dinosaurs with pennaceous feathers are fliers or 
probably neoflightless, there is no evidence that such complex structures 
evolved outside the context of flight. On the other hand, simple feathers 
evolved before flight, and it is possible they first appeared in fairly basal 
archosaurs and were inherited by pterosaurs, protodinosaurs and dinosaurs. 

Since at least some basal dromaeosaurs were better fliers than Archaeopteryx 
the conventional hypothesis that dromaeosaurs are preavian nonfliers or 
protofliers is dead, dead, dead. It is as falsified as the now silly notion 
that birds are not dinosaurs, or the hypothesis that humans were specially 
created by a loving god a few thousand years ago. The burden is now upon 
those who wish to revive the conventional scenario over the thriving 
neoglightless alternative. Good luck. I used to say that the question of bird 
descent from dinosaurs has been settled, and that the question is now how 
birds evolved from dinosaurs. To this I add the question is not whether some 
dinosaurs were fliers and neoflightless, the question is the details of the 
process. The neoflightless dino-bird revolution has occurred, get used to it. 
Dromaeosaurs were sophisticated fliers (some or all basal examples) or 
secondarily flightless (derived examples). They are very probably closer to 
modern birds than Archaeopteryx (see DA for the analysis). Since troodonts 
are probably relatives of dromaeosaurs, and basal troodonts are more 
bird-like than derived examples, they too should be neoflightless 
post-urvogels. The evidence that caudipterygians and other oviraptorosaurs 
were neoflightless post-urvogels, and perhaps even derived avebrevicaudians, 
is growing. May be true of therizinosaurs as well. The number of reseachers 
supporting neoglightless dinosaurs is balloning, a classic sign of a paradigm 
shift. Instead of being the neat and tidy progressive development of flight 
from ground running dinosaurs to ever more capable flying birds imagined by 
cladists, the origins of birds and their flight was a typical evolutionary 
hodge podge of some creatures going up and other down, with the basal avian 
radiation throwing off a blizzard of secondarily flightless dino-birds. 

Steve may be polemical about it but so what, because he is correct that the 
cladistic cult - in which a methodology that is a merely one of a set of 
phylogenetic tools has widely been taken to be to the only legitimate means 
of restoring phylogeney - has been dealt a grevious and well deserved blow. 
It is now up to the cladists to try to explain how the methodology managed to 
ignore or deny the excellent evidence for dinosaurian neoflightlessness for 
two decades. The descent of humans from derived apes was demonstrated without 
cladistics (remember Darwin?), the descent of mammals from derived therapsids 
was demonstrated without cladistics, and Ostrom demonstrated the descent of 
birds from derived theropods without cladistics. Cladistics has its uses and 
achievements, but it is inherently limited. Cladistics simply cannot handle 
the massive reversals inevitably associated with flight loss in birds so 
basal they are still basically dinosaurian in form. It should never again be 
required that a phylogenetic study be required to include a cladistic 
analysis. It should be a matter of researcher preference. Doing otherwise 
violates academic freedom. Never should cladograms be taken at face value, or 
considered to automatically prempt alternative methods (this does not mean 
that other methods automatically trump cladistics either, each situation and 
method has to be objectively examined and weighed on a case by case basis). 

Think about it. According to the common cladistic view, once upon a time in a 
fantasy Mesozoic, post-Archaeopteryx birds initially evolved in accord with 
some mystical, progressive force to improve flight so strong that they 
somehow failed to produce neoflightless forms until such advanced birds as 
hesperornithiforms came along. This is a gross violation of basic 
evolutionary principles, in which there is no progressive force. Things often 
go into reverse when it is possible to do so. Obviously, the first fliers 
would actually be especially prone to losing flight, neoflightless dino-birds 
should have been running all over the Cretaceous and people should have been 
looking for them. I showed that a number of Cretaceuous avepods clearly were 
chock full of neoflightless characters back in the 80s. What is disturbing is 
how few others even tried to look, and how the cladocult ignored the reality 
of evolution and the osteological evidence because the computers kept 
spitting out the same results that the inflexible methodology could not avoid 
doing. It's not the cladograms' fault, it's the fault of those who took them 
too seriously. It was so bad that at times I was forced to take the ossified 
uncinate processes off my dromy skeletons even they they are clearly present 
in the quarry photo of the fighting Velociraptor! When cladograms suggested 
that the entire array of Cretaceous avepectoran dinosaurs were not 
neoflightless, leaving no examples of basal secondarily flightless birds, 
then the logic of cladistics should have been questioned. Instead cladists 
have been writing papers stating that pennaceous feathers have definitely 
been shown to have evolved before flight, when the only dinosaurs that have 
such feathers always show a set of avian flight features (very large ossified 
sterna, ossified sternal ribs and uncinates, flattened central fingers or 
reduced outer fingers, pterosaur-like or abbreviated tails) more derived than 
those seen in Archaeopteryx. The folding arms of dromaeosaurs etc did not 
evolve as the cladists imagined as preflight snap action devices - which are 
limited to sit-and-wait predators - they evovled to fold up wings, duh. Good 
science never gets trapped in a single methodology, it always uses multiple 
approachs. The obsession with cladistics has been a major failure of the 
science of paleophylogeny, and set back our understanding of the origin and 
early loss of avian flight. But we're catchin up. 

What is more important for phylogentics than cladistics? Simple. Fossils. 
Especially transitional fossils. Cryptovolans tells us far more about the 
origin and nature of dromaeosaurs than all the cladograms yet produced.

I'll be blunt. We have little idea how the assorted avepectoran dinosaurs are 
related to one another in a detailed manner. The degree of mosaic evolution, 
convergence, parallalism and reversal is so massive - look at 
Scansoriopyteryx with it's hodge-podge of characters from assorted groups 
(but the short, vertical pubis is found in derived troodonts and does not 
indicate a basal position) - that we simply do not have a system that can 
restore the relationships. You can do the most sophisticated cladograms, mix 
them all together and put them through a shredder and you'll be no worse off. 
The only solution is more transitional fossils. 

And while we're at it, I've had it with the obsession over the reversed 
hallux and toe spread So what? The only important thing about the hallux is 
that is exists, and that it may have evolved only once, in the Avepoda. Since 
the 1st metatarsal is not attached proximally, the toe is free to point 
whatever way it happens to for adaptative needs. So Scansoripteryx, 
Archaeopteryx and most birds have it, and most dino-avepods, kiwis and 
shorebird halluxes are not reversed. Both cladistic and antidinosaur analyses 
that cite this trivial attribute as an avian-only feature are bogus. Toe 
spread including hallux reversal is a functional attribute that could and 
should have evolved in shore dwelling dinosaurs that walked on squishy mud. 
So what there are lots of bird-like tracks in the Triassic and Jurassic? Just 
means there were lots of Avepoda with bird-like feet, not that they are birds 
or even especially close to birds. 

Instead of nitpicking the papers in the Czerkas volume, professional 
paleontologists would be better advised to take a hard look at their own 
methods. Over the last few decades it has been paleos on the edges of the 
community that have done the most to advance the field. It was Bakker who 
pushed the concept that sauropods were terrestrial high browsers, that 
dinosaurs form a united group, thay dinosaur energetics was more bird than 
reptile like, and that small avepods were feathered. Ostrom showed that birds 
where dinosaurs without cladistics. Moi pushed feathered dinosaurs, showed 
that some dinosaurs were secondarily flightless, and proposed the only 
logical and empirical explanation for giant animal energetics, 
terramegathermy. Czerkas did the careful reexamination of a previously publi
shed specimen that showed a sickle clawed dinosaur did fly. High school grad 
Horner showed that some dinosaurs took care of their wee babies whihc then 
grew up fast. The abundant research of academics has done an important job of 
tending to verify the radicals. There has been and always been a need for a 
blend of researchers from the usually conservative academics to the rads 
without official credentials. Forcing everyone to toe the process and 
credentials line would be a big mistake. Paleo would be a decade or two 
behind where it is now. That the Czerkas book does not comfortably fit into 
what academics want is not important, it is what it contains that is. 

G Paul