Re: Digit Loss

["Steve Brusatte" <dinoland@lycos.com>]

On Thu, 07 Jun 2001 19:05:11  
 Daniel Bensen wrote:

I wrote:
>>> What is interesting is that some theropods have lost two digits, and these 
>>> digits never seem to re-evolve.  This may mean that the evolutionary 
>>> pressure still existed, preventing the re-evolution of the primitive state. 
>>>  <<

Dan wrote:
>    I disagree here.  Remember that evolution is driven by random mutation.  

Yep.

>To loose a digit or limb is a relatively simple mutation: fail to turn on a 
>switch during development, and the embryonic digit is starved for nutrients, 
>or the proto-digit cells fail to propagate, or some other pathway is blocked.  

Yeah.  Larsson gave an excellent talk on this at the Armour Symposium, as those 
of you who were present know.  I already summarized his talk awhile back 
onlist, but he basically said that there are two factors in digit loss: loss of 
the digit itself (loss of the chondrocytes/osteocytes), and a loss of the digit 
position (before the cartilage forms).  Very interesting stuff.

>To create an entire digit out of whole cloth, however, is quite another 
>matter. Once a gene is deleted, it's gone.  Well, with all the redundancies of 
>the genome, primitive genetic material may exist somewhere, but the >>series 
>of mutations necessary to bring an old sequence back into expression is 
>probably no easier than
>creating the genes from scratch.  

That's a good point, but we must remember that dinosaurs, especially theropods, 
had millions of years to evolve, re-evolve, etc.  The frill and horns of 
ceratopsians were new structures, yet they evolved.  Horny beaks evolved in 
some ornithopods.  The same is true for the plates of stegosaurs.  These were 
new structures in many ways, and they evolved.  It took millions of years, but 
it happened.  Why wouldn't digits (or limbs) re-evolve in dinosaurs?  This 
tends to make me think that something was holding the re-evolution of digits, 
at least in theropods, back.  Maybe it was the evolution of the wing or 
locomotion, as some listmembers have suggested.  Personally, I think the answer 
may lie somewhere else.  

>A random mutation, falling anywhere in the genome, is much more likely to 
>destroy a digit than to create one.

Yes.  Good point again.  This is true for a single mutation.  However, when an 
organism or a lineage of organisms has millions of years to evolve, I am sure 
that somewhere along the line the odds will say that a mutation promoting the 
re-evolution or creation of digits will occur.  After digits have been lost 
we're back at ground zero.  Re-evolving digits is like evolving any new 
structure, whether it be a frill or a beak.  New structures commonly turn up.  
Why couldn't digits re-evolve?

>    We see numerous examples of digit-loss in tetrapod evolution, but no digit 
> creation.

Why?  I don't know.  The earliest tetrapods, like Ichthyostega (sp?) had 
something like eight fingers.  Digits were created in this genus.  However, 
over time it seems tetrapods, especially dinosaurs, seem to lose these digits.  
Dinosaurs, as a clade, have gone from 5 down to 0 (in some cases) digits 
throughout their evolution.  Once again, it seems to me as if something, some 
evolutionary pressure, is preventing the re-evolution of digits.  A mutation 
for the re-creation of digits must have occurred at some point in the history 
of dinosaurs, yet no theropods, at least, re-evolve digits.  Why?

>    For most animals, digit-loss would represent a neutral mutation, or 
> perhaps a mutation of slight benefit (as nutrients that might have gone to 
> nourish a lost digit can now be diverted elsewhere), so, in most cases, a 
> mutation that causes the loss of a digit should be preserved, not because of 
> any particular advantage of a small number of digits, but because of the 
> general sloppiness of natural selection.  

Would the mutation really be neutral?  I would tend to say that the mutation 
would offer an advantage.  Since digits don't re-evolve, there must have been a 
disadvantage to retaining these digits.  This disadvantage probably wasn't 
simply something that was neutral.  True, an animal that re-evolves a neutral 
mutation would not be any more successful, and would not produce any more 
offspring, than an animal without that neutral mutation.  However, wouldn't the 
re-evolution of digits or the lengthening of shortened limb bones be successful 
for some animals?  In humans, our five fingers (primitive) were likely retained 
for hunting.  It would be dangerous for humans to lose these digits.  Imagine 
trying to throw a spear (or a baseball...or type!) with only four digits, or 
three.  But, in dinosaurs it looks to me as if it was beneficial to lose these 
digits, and dangerous to retain them.  Why?  I don't know.  

>    To get to the point, digit number in dinosaur clades represents a series 
> of trivial mutations, responding to very weak selective pressures, just the 
> sort of thing that is so useful in classification.  As Darwin said in _The 
> Origin of Species_ (I paraphrase because I can't seem to find my copy), these 
> trivial details (openings in the skull, bumps on molars, or number of toes), 
> respond to no particular selective pressures and so vary more or less 
> randomly through time, making them useful in determining phylogenies.  We 
> cannot safely classify dinosaurs based upon the shape of their claws (as the 
> selective pressures that
>govern these implements are strong), but we can use number of digits.

I agree with much of that, but I do think that selective pressures that govern 
digit numbers are strong, too.  Why don't digits re-evolve in theropods?  I 
realize that re-evolving a digit is like evolving a new structure, and is 
unlikely in the short term.  However, theropods had millions of years to 
evolve.  Why is there no case of a theropod re-evolving a digit?  There had to 
be some sort of pressure.  Re-evolving digits must have been harmful.

Whew!  Now I want to sit down and relax!

Steve

Steve Brusatte
Dino Land Paleontology
http://www.geocities.com/stegob


Get 250 color business cards for FREE!
http://businesscards.lycos.com/vp/fastpath/