Adaptive advantage (was Re: ABSRD BAND on Sinornithosaurus feathers)

["Tim Williams" <twilliams_alpha@hotmail.com>]

A few words regarding this confused babble...

>    The model of feather development proposed by Prum, and illustrated
> in fig. 6 of this paper, also makes no adaptive sense. If every stage in 
> the
> evolution of a morphological structure has to be adaptive, then it is
> difficult to fathom how a simple hollow cylinder or a pine needle-like 
> branching structure can be functional in an animal.

And again:

> It is therefore
> foolhardy to fly with wings made from interelocking filaments, not knowing 
> a priori whether the interlocking structures are strong enough or 
> structurally sound enough to serve functionally as wings.

OK, I think I know what they're getting at.  Suchs assertions (which come 
thick and fast from the ABSRD/BAND camp) stem from the misguided premise 
that the purpose of feathers FROM THE VERY BEGINNING was FLIGHT.  In other 
words, feathers first evolved as aerodynamic structures, and every 
modification thereafter improved the ability of the feather (and, by 
extension, the wing composed of these feathers) to provide the animal with 
lift.

Well, not necessarily.  Most likely, not at all.  The branching structures 
we in _Sinornithosaurus_, _Microraptor_ etc may have initially served in 
some function that had nothing to do with aerial locomotion: 
thermoregulation or display have been suggested.  (By Richard Prum, among 
others.  Gosh, how 'bout that.)  These branching, non-aerodynamic feathers 
may have served as wonderful heat-trapping structures.  There's a novel 
idea!

The function of incipient structures is something that confounded Darwin, 
and it still bedevils evolutionary biologists and paleontologists today.  It 
doesn't befuddle scientists as much as it used to though.  The notion of 
"exaptation" - that anatomical structures are evolved for one purpose, and 
later in evolution are selected for a totally different purpose - has been 
around for quite a while.  A certain property of the structure in question 
pre-adapts this structure to performing another function.  (If memory 
serves, Gould and Vrba actually coined the term "exaptation" in 1982; but 
the very concept is older than that).

I like the notion that the feathers evolved first for insulation, then took 
on a function in promoting drag in small facultatively arboreal theropods 
that leaped from tree branches onto prey below.  Essentially this is the 
"Pouncing Proavis" model of Garner, Taylor and Thomas (except that it makes 
better sense to put the proavian up in the branches of trees, like 
_Microraptor_, rather than having to jump from rocks.)  The original purpose 
of the "proto-wing" was drag.  Having drag-inducing devices positioned far 
from the body's center of gravity (such as on the hands and the end of the 
tail) allows improved maneuverability.  This is an excellent adaptation for 
a leaping predator which first sights the prey from the vantage of a tree 
branch, then endeavors to (1) take prey by surprise and (2) land as close as 
possible to the prey.

Here we come to another flaw in the Feducciary model.  ABSRD/BAND proponents 
are certain that the ancestors of birds were gliders.  Gliders come as two 
types.  In the first type, the aim is to increase lift (i.e. high 
lift-to-drag ratio) by developing a wide lift surface across the body.  This 
is what we see in modern gliding mammals - such as "flying" squirrels, 
"flying" possums (phalangers), and "flying lemurs (the colugo - not a true 
lemur), which have a patagium stretched between the fore- and hindlimbs 
contiguous with the body wall.  The added lift allows these animals to 
extend the duration of aerial leaps - useful for getting from one tree to 
the next by avoiding the ground below.

In these creatures, gliding efficiency (measured as the horizontal distance 
travelled) is increased.  This is probably the rationale behind the putative 
gliding structures of an array of arboreal Permo-Triassic reptiles: 
kuehneosaurids, coelurosauravids, _Sharovipteryx_, and perhaps _Longisquama_ 
(assuming the appendages were paired and could be spread out laterally to 
catch the air).  In all these creatures the gliding surface extends from the 
body.  (_Megalancosaurus_ was probably not a glider, as Ruben has claimed - 
see Silvio Renesto's work on this genus.)  ABSRD/BAND supporters are 
currently scouring the Permian and Triassic fossil record for gliding 
reptiles that show such adaptations.  Ignoring (for a moment) that there is 
overwhelming support for the origin of birds from among theropods, their 
quest is doomed to failure on biomechanical grounds as well.  I'll mention 
why later.

The other type of gliding is called "gliding" but strictly speaking it isn't 
true gliding.  It's often referred to as "parachuting" or a "controlled 
fall".  The aim is to increase drag, not lift (although this may become an 
important component later).  By having drag-creating surfaces, the animal 
improves its maneuverability in descents from an elevated surface (such as a 
tree branch) to a lower surface (such as the ground).  In parachuters, the 
drag-creating surface is positioned FAR FROM THE BODY.  The maximizes 
maneuverability in the air, but results in poor gliding efficiency.

This is what "flying" frogs do (e.g. _Rhacophorus_).  Contrary to what I 
have read in at least one book, the webbed hands and feet of flying frogs do 
not act as gliding surfaces.  When jumping from a tree branch, the webbed 
hands and feet DO NOT increase the horizontal distance travelled in the air 
by the frog.  In fact, the webbed surfaces may actually decrease gliding 
efficiency.  (This has been determined experimentally using both models and 
living frogs - and both "flying" frogs and closely related non-flying 
frogs).  The webbed surfaces, which are positioned far from the body, 
promote drag, allowing greater maneuverability during descents (good for 
dodging leaves and tree branches on the way down) and precision in landing.

Sifakas (_Propithecus_) are arboreal mammals (lemurid primates) have a mat 
of hair on the trailing edge of their arms, which may serve a similar 
purpose when jumping from tree to tree.  (Feduccia has discussed these as a 
modern analog for a proavian - not realising that it undermines his arboreal 
glider model for the origin of flight in birds.)  Interestingly, in both 
flying frogs and sifakas there are NO obvious SKELETAL adaptations for this 
lifestyle.  The drag-creating surfaces are developed entirely from the 
integumentary structures: skin or hair.  This is why the notion of looking 
for Permo-Triassic reptiles that show both parachuting adaptations and 
bird-like characters is so wrong-headed (apart from the 99 other reasons).

Let's look at _Archaeopteryx_.  It shows primaries on the manus, secondaries 
on the ulna, but no remiges on the humerus.  No tertiaries have been 
identified in any _Archaeopteryx_ specimen: the inner arm was devoid of 
flight feathers.  Also its tail is long and covered in a pairwise 
arrangement of retrices.  Although the tail feathers are asymmetrical, the 
long wide tail is not much good for generating lift.  Excellent for creating 
drag though.

Let's look at _Caudipteryx_.  It has primaries on the manus, but no 
pennaceous feathers (remiges) on the forearm or humerus.  It has retrices 
(paired and laterally-splayed as in _Archaeopteryx), but these are 
restricted to the distal half of the tail.  What's more, the symmetrical 
feathers are excellent for producing drag when oriented perpendicular to the 
air flow (as they would be during leaps to the ground).

_Protarchaeopteryx_ also has retrices at the end of the tail.  The forelimbs 
do not show evidence of feathers though (I'm not certain if this is due to 
absence in the living animal, or lack of preservation.)

The adaptive advantage of stiff, vaned, non-asymmetrical feathers is clear: 
drag and consequent maneuverability.  This was the primordial force which 
eventually led to powered flight.  This explains why the incipient flight 
surface (evident in the "proto-wings" of _Caudipteryx_) evolved distally to 
proximally in both the forelimb and tail.  The incipient flight surface did 
not first appear close to the body wall (at the armpit) nor equally along 
the entire length of the forelimb+manus - the very arrangements one would 
expect in a glider trying to maximize lift.

_Microraptor_ shows us that small maniraptorans (deinonychosaurs, no less) 
could climb trees.  They may not have spent their entire time there: they 
were facultatively arboreal, comfortable both in trees and on the ground.  
Same for _Archaeopteryx_.  The above model for the origin of avian flight is 
much like that proposed by Sankar Chatterjee in _The Age of Birds_, without 
the incipient lift surfaces or the unnecessary diving stage.

_Caudipteryx_ may not have been at all arboreal, but its ancestors probably 
were.   Basal oviraptorosaurs and basal deinonychosaurs (like _Microraptor_) 
could have been sem-arboreal; Sereno gives the Oviraptorosauria and 
Deinonychosauria as sister groups, so the common ancestor may have been 
scansorial and at least partly aboreal.  The obvious terrestriality of 
_Caudipteryx_ and oviraptorids, and troodontids (assuming a monophyletic 
Deinonychosauria) and velociraptorines, would be secondary under this 
scenario.  Chatterjee also suggests that the opisthopubic pelvis and 
semilunate carpal may also be scansorial/arboreal adaptations; both ideas 
deserve further investigation.  (I prefer to regard the semilunate carpal as 
a pre-adaptation for tree-climbing.)  With this scenario in mind, I would 
like to take a closer look at the claw geometry of _Microvenator_.

Further, the major anatomical innovations between basal paravians (of 
Sereno) or eumaniraptorans (of Holtz) and _Archaeopteryx_ probably occurred 
entirely in the integument.  There is not much difference between the 
skeletons of _Sinornithosaurus_ and _Archaeopteryx_, but I'm sure a 
helluvalot of change occurred in the arrangement and structure of the 
feathers.

Finally:

> It is quite obvious that the cladists are influenced primarily by their 
> cladograms, rather than by the structural similarities between avian 
> feathers and Longisquama feathers.

Guilty as charged.  What audacity paleontologists have, claiming birds 
evolved from theropods simply because they're closely related.  Cheek!

Anyway, that's my thoughts.  Apologies for the long-winded post - 
considerably more than 2c worth.


Tim

------------------------------------------------------------

Timothy J. Williams

USDA/ARS Researcher
Agronomy Hall
Iowa State University
Ames IA 50014

Phone: 515 294 9233
Fax:   515 294 3163

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com