NO SECONDARILY FLIGHTLESS THEROPODS

["Matthew Troutman" <m_troutman@hotmail.com>]

     Recently, there has been a lot of talk about secondarily flightless 
theropods, most specifically dromaeosaurs. I was once a believer in this 
theory as modeled by GSPaul in '88 ( PDW, a classic ). But, as I have 
become more interested in the mechanisms of flightlessness in modern 
birds, I have discovered that many of the features in the supposed 
"flightless" theropods are not actually secondarily flightless 
characters, but can be explained by other means.

For those interested in history, a strange sort of "flightless" theory 
was proposed by Lowe in 1935. He proposed that all of the flightless 
ratites and penguins never really did fly, but were rather direct 
descendents of small theropods such as Ornitholestes. Simpson refuted 
and disproved this theory siting Archaeopteryx, a pygostyle, apteria, 
and other volant characters as evidence that these animals once flew. 

In 1988 Gregory S. Paul fully outlined his theory ( foolowing a 1984 
version ) that some of the more derived theropods were flightless 
descendents of an Archaeopteryx-like ancestor(s). The foremost example 
of this, according to him, were the birdlike dromaeosaurs, which had 
shoulder girdles like some flightless birds and several other volant 
adaptations. Also, they appeared to be more advanced than Archaeopteryx 
in certain features. 

George Olshevsky furthered the framework of Paul's conclusions to 
suggest that not only that some theropods were flightless, but that all 
dinosaurs had arboreal birdlike ancestors. ( See Olshevsky 1991, 1994 
and posts to this list ). 

However, as compelling and intriguing as these hypothesis are, they do 
not fully appreciate the fundamental bauplan that flightless birds have. 
The unifying character between flightless birds is that they all show 
the same consistent paedomorphic trends ( visible sutures, atrophy of 
the forelimb, massiveness of the vertebrae and hindlimbs, large orbits, 
etc ). The lack of many of these features in "protoavians" ( for lack of 
a better term ) would seem to argue against secondarily flightless 
theropods. 

Before I look at the "cases", let me reiterate the points I have made in 
previous posts to the list :

Paedomorphosis is the mechanism that makes flightless birds flightless. 
It is seen in all flightless birds that we know to date. Indisputedly, 
this is the feature that makes birds flightless and it has been proven 
in laboratory studies where the thyroid was taken out of juvenile birds 
and they began to exhibit retarded development in the form of the 
feathers, forelimb, shoulder girdle, hindlimbs, and behavioral changes. 
How these paedomorphic trends work in birds is when species gets 
isolated ecologically, geographically, or both they get turned to 
selective advantage. Usually birds turn flightless on islands or areas 
with few predators, so flight is not necessary. Taking of new ecological 
niches has also been considered a compelling reason ( See Feduccia 1996; 
Dawson et al. 1994; Olson 1973. for further review on the points above).

This biogeographic isolation would be unlikely to happen in the Late 
Jurassic or Early K because Pangaea was just beginning to break apart 
and there were few islands to get isolated on. Ecological isolation may 
have been unlikely, but there is no evidence that we know of now. 

The only "flightless" birds that do not show paedomorphic features are 
the penguins and auks. But, since these birds are essentially still 
"fliers" ( have anyone ever seen a penguin " fly " through water? ) 
there should be no reason why all of the typical flightless bird 
characteristics ever evolved. These birds will be excluded from this and 
when I say flightless birds I mean the typical flightless birds.

Now to the cases : ( Read Paul 1988 and posts to the list arguing these 
topics ).


Dromaeosaurs are the supposed "foremost" example of secondarily 
flightless theropods. GSP has expanded this notion very much based on 
the similiarities between flying Archaeopteryx and flightless 
dromaeosaurs. The two groups do share a buttload of features including 
but not restricted to : 

1) Dorsally depressed nasal.
2) Diamond shaped supraoccipital.
3) "Twisted" paraoccipital process.
4) Parallelogram-shaped ilium.
5) Large and reversed pubic peduncle.

GSP concluded that dromaeosaurs, since they are more "advanced" and 
share a similiar shoulder structure with flightless birds. This 
hypothesis was expanded when a recent list member called the "Rahona" 
bird a link between Archaeopteryx and dromaeosaurs. 

However, these two conclusions have many faults. First of all, 
dromaeosaurs have no features that can be considered more advanced than 
Archaeopteryx ( except for perhaps the sternal size and pubic 
retroversion ). All of the best recent finds of dromaeosaurs and 
Archaeopteryx suggest that Archaeopteryx is more birdlike than all known 
dromaeosaurs. It shares with modern birds : 

a) Triangular teeth with no serrations and replacement pits that are 
closed ventrally ( similiar to crocodilian teeth ).
b) Lack of postorbital-jugal contact.
c) Loss of squamosal-quadratojugal bar.
d) Quadrate with orbital process.
e) Jugal bar.
f) Coracoid that faces totally posteriorly and forms an acute angle with 
the scapula.
g) Single sternal element ( as opposed to two sternal plates as seen in 
dromaeosaurs ).
h) Anterodorsal ischial processes.
i) Reversed distal hallux.
j) Caudal vertebrae count less than 23.
k) Triradiate palatine.

With these features which Archaeopteryx shares only with early birds, it 
becomes clear that Archaeopteryx is more birdlike than dromaeosaurs. If 
dromaeosaurs are derived from Archaeopteryx massive reversals would have 
to take place and dromaeosaurs would be clasified as birds. 

"Rahona" shows many dromaeosaur-like features, with a few avian features 
thrown in. However, it is unlikely that it is ancestral to dromaeosaurs 
because :

i) The glenoid faces laterally ( in flightless birds the glenoid still 
faces somewhat laterally ).
ii) Ischial processes are still present.
iii) Radioulna is elongated ( explained below ).
iiii) Reversed hallux is present.

Let me explain iii. In flightless birds the forelimb remains 
consistently short ( less than 50% of the hindlimb ), even in 
phorusrhacids. If dromaeosaurs are secondarily flightless, than their 
long forelimbs (75% and above ) would be a unique feature.

The shoulder structure of dromaeosaurs was the paramount feature that 
allowed GSP to conclude that dromaeosaurs were secondarily flightless. 
However, this does not tell us much, the dromaeosaur shoulder is more 
strongly built than the typical flightless bird shoulder. Plus, it shows 
an angle that is in between 90 degrees and 100 degrees, whereas the 
typical flightless bird shoulder is 180 degrees to 140 degrees.

The sternal structure in dromaeosaurs is also different from that of 
birds. It is two sternal plates in dromaeosaurs, in contrast to the 
single unit in birds. In most flightless birds the sternum remains a 
single unit (Thambetochen, the Hawaiian flightless goose, has a sternum 
that is unfused and two plates in the posterior half though ). ( Olson 
and Wetmore 1976 ).

The folding nature of the forelimbs was sited as a flight 
characteristic, but it could have evolved for many reasons ( keep 
feathers out the manus, balance, posture,etc ). 

In summary, dromaeosaurs simply do not show the typical trends that 
typical flightless birds exhibit. They do not show the paedomorphic 
features that all flightless birds have. The supposed "secondarily 
flightless" characteristics are not congruent with what is seen in 
flightless birds. Many of the volant characteristics that are seen in 
dromaeosaurs can be easily explained as scansorial characteristics (see 
Chatterjee's book ) and are simply not in line with volant ancestry. 

Oviraptorosaurs, troodontids, ornithomimosaurs ( and occasionally 
tyrannosaurs )  have all been part of the secondarily flightless theory. 
However, many of the same features that are supposed evidence of volant 
descent are the same as dromaeosaurs. What was the main reasoning behind 
the idea that these animals are also secondarily flightless is that they 
show some features that are supposedly more "advanced" than 
Archaeopteryx and dromaeosaurs. Many of these features are braincase and 
dental features. The arrangement of the VI nerve, the braincase 
pnematicity, encephalization ratio, and the form of the teeth, are among 
these features. However the VI seems to be in a birdlike position in 
dromaeosaurs ( as hinted by Norell in the Audoban article ), the 
braincase pnematicity is not unique among "protoavians", and the dental 
morphology ( dubbed "troodontiform" for therizinosaurs, troodontids, and 
ornithomimosaurs ) though birdlike in the root morphology, differs from 
that of birds. The only thing that is totally unique is the 
encephalization ratio, though some dromaeosaurs purportedly have large 
brains. 

Lets deal with the main groups individually:

Oviraptorosaurs show a similiar shoulder morphology and similiar sternal 
morphology to dromaeosaurs. However, the same arguments that I have used 
for dromaeosaurs apply to oviraptorosaurs. Oviraptorosaurs also have a 
birdlike humeral morphology with prominent deltoid crests and a similiar 
furcula morphology to early birds. However, in all flightless birds ( as 
noted above ) the forelimbs atrophy to thin vestiges ( phorusrhacids are 
the derived exception where their short forelimbs were architecturally 
refined to a point not seen in theropods ) and the furcula turns into 
either an obtuse-angled vestige or degenerates into two splits. Other 
than these features, oviraptorosaurs show no other remarkable 
"secondarily flightless" characters.

Troodontids and ornithomimosaurs will be treated the same here. Of the 
two groups, troodontids show the most birdlike characters. They have a 
similiar shoulder angle to dromaeosaurs and oviraptors. 
Ornithomimosaurs, however, do not show the same features and all of the 
arguments used above apply here. Plus, using tyrannosaurs as an 
outgroup, the birdlike features are derived.

One thing that argues against the volant ancestry of these groups is the 
propubic pubis. A retroverted pubis would be better for flying ecause it 
makes a airfoil ( convex dorsally and concave ventrally )  shape of the 
body. A propubic pubis would create a reverse-airfoil, which would not 
be advantageous for flying because it would not create lift and the 
"bird" would divebomb to the ground! Paul ( 1988 ) argued that the 
propubic pubis was a reversal, which it may well be, but there is no 
reason why that these groups should evolve a propubic pubis. 

Several listers have stated that, " Since dromaeosaurs had teeth and 
seperate fingers, then the paedomorphosis argument doesn't apply here." 
This does not hold up when phorusrhacids are taken into account. 
Phorusrhacids have rather short forelimbs but they are structurally 
refined ( really strong and robust ) and the wrist lacks the 
characteristic avian wrist flexion. This is a predatory flightless bird 
forelimb. This also does not hold up when the consistentcy that 
paedomorphosism makes birds flightless and still shows up.

Another argument used against this is, " A lot can happen in tens of 
millions of years of evoltution." However, this does not take account 
that the ratite's ancestors, the lithornithids, are Paleogene in age and 
ratites have been around roughly coeval to the lithornithids. 

Well, hopefully this should show the shortcomings of the secondarily 
flightless scenario. For those who actually read this : thanks for 
staying with this! Any thoughts, comments, faults, and congratulations 
are welcome.

Here's the list of refs:


Paul, Gregory S. Predatory Dinosaurs of the World. 1988.

Lowe, Percy R. 1935. On the relationships of the Struthiones to the 
dinosaurs and to the rest of the avian class, with special reference to 
the position of Archaeopteryx. Ibis, 13th series., 5:298-429.

Olshevsky, George. Mesozoic Meanderings 2.

Olshevsky, George. Omni, June, 1994.

Feduccia, Alan. The Origin and Evolution of Birds. 1996.

Dawson et al. 1994. Ratite-like neoteny induced by neonatal 
thyroidectomy of European starlings, Sturnus vulgaris. Journal of 
Zoology ( London ) 232: 633-639.

Olson, Storrs L. 1973. Evolution of the rails of the South Atlantic 
islands ( Aves: Rallidae ). Smithsonian Contributions to Zoology 
152:1-53.

Olson and Wetmore. 1976. Preliminary diagnoses of extraordinary new 
genera of birds and Pleistocene deposits in the Hawaiian islands, 
Proceedings of the Biological Society of Washington 89: 247- 258.

Regards and thanks,


Matt Troutman

______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com