Path: utzoo!utgpu!water!watmath!clyde!att!rutgers!ucsd!ucbvax!pasteur!cory.Berkeley.EDU!jyamato
From: jyamato@cory.Berkeley.EDU (YAMATO JON AYAO)
Newsgroups: sci.bio
Subject: Re: Evolution
Keywords: evolution
Message-ID: <4240@pasteur.Berkeley.Edu>
Date: 3 Jul 88 22:28:38 GMT
References: <5944@bloom-beacon.MIT.EDU>
Sender: news@pasteur.Berkeley.Edu
Reply-To: jyamato@cory.Berkeley.EDU.UUCP (YAMATO JON AYAO)
Distribution: sci.bio
Organization: University of California, Berkeley
Lines: 59

In article <5944@bloom-beacon.MIT.EDU> orwant@athena.mit.edu (Jonathan L Orwant) writes:
>I'm a bit confused about how special-purpose anatomical parts can evolve, e.g.
>the evolution of a bird's wings from an unwinged predecessor.
>If it happens gradually, then why isn't the 2% of a wing along the way an
>evolutionary detriment?  I've heard that it acts as a heat insulator; maybe 
>so, but I would think that there would be much more efficient ways of 
>retaining heat that would give competing organisms an advantage in the 
>meantime.

The general feeling is that there is often some function served by
the precursor organ, often one which is not apparent when examining
the final evolutionary result.  For example, flaps of skin along the
limbs and sides (the apparent precursor of wings in mammals, at least)
can be useful for gliding, keeping infants warm, storing fat during
hibernation, lots of things.  Finding out what the ancestors of birds
used their modified limbs for could be tricky, but the answer may be in
the fossil record.

It is also possible for moderately complex structures to occur as a more
or less random result of changes elsewhere in the genome.  Consider the
human big toe.  We may have well-developed big toes because we once
needed them, but it is also reasonable to suspect that the same programming
used to do hands is being used to do feet.

A poster to talk.origins recently brought up the suggestive if not
conclusive finding that insect wings (harder to explain than bird wings,
since it can be shown genetically that they are *not* modified legs) are
functional as heat exchangers, and are in fact used this way by butterflies.

>Another explanation I have heard is that specialized master-genes turn on
>entire sequences of genes so that once a mutation triggers the proper master
>gene, an entire wing appears.  If this is the case, where is the information
>of how to grow a wing stored in an animal whose ancestors were wingless?

>-Jon Orwant (orwant@wheaties.ai.mit.edu)
>Pity me.  I'm a computer scientist. 
You're right, this is not a plausible explanation for the appearence of a
whole, specialized wing *de novo*.  What I think is being got at here is that
a program normally used for one purpose can be accidentally triggered at
a different time or place or under different circumstances, resulting in
a fairly complex new structure (because of its interactions with its
surroundings).  For example, fruit flies can go from having two wings
to four in a single mutation, or from having antennae to having
legs on their heads--the information is there already.  In theory perhaps
the leg-headed flies could eventually develop some use for those
legs (in fact, this could be the origin of antennae--a misplaced leg
subroutine).

Computer science is not a bad starting point for understanding
genetics.  The development of the nematode is my favorite example--
the entire pattern of cell division, differentiation and death to
produce an adult nematode (1100 cells) is known, and there are some
subroutines clearly visible--patterns which repeat in many parts of the
program.  There are also some nice switches, where one minor variable
controls expression of large blocks of code.

Mary Kuhner
graduate student, genetics, UC Berkeley
(but my opinions are my own)