Tim Tyler  wrote in
news:c28lqm$13er$1{at}darwin.ediacara.org: 

> Jim Menegay  wrote or quoted:
> 
>> 2. What does it mean for one species to be more fit than another
>> species? Two possibilities.  One (mentioned by Fisher - I just read
>> this today) is that a fit species is one that minimizes its risk of
>> extinction.  The second (championed by Gould, I think) is that a fit
>> species gives birth to a multitude of descendent species.  Either
>> formulation makes sense - but neither is likely to be subjected to a
>> Popperian test.  These are ideas for modeling only.
> 
> Fisher's idea is good - provided he extends the concept to any
> descentants of the species.

Some random thoughts (and I think all my thoughts on this topic are still 
pretty random):

There apparently is research (see reference in 13 Feb 2004 Science p. 
973) that speciation is more rapid in polyandrous species,apparently  as 
a result of greater sexual conflict leading to more rapid evolution of 
reproductive traits. But does this make the species more fit?

On a population ecology level, many species exhibit density-dependent 
reproduction. This may help stabilize a species from extinction due to 
wild population swings, but only some of the mechanisms (migration, 
increased territoriality) appear to have an individual benefit. Others  
(stress shock, infant mortality) would only be valuable at the species 
selection level.

Gould's idea is interesting. If genera that give rise to many species 
have a longer lifespan than genera that don't, then mechanisms that give 
rise to speciation will be selected (although I have difficulty calling 
such mechanisms "adaptations", because they will for the most part be 
otherwise selectively neutral.) There was an interesting article in 
Science several years ago (apparently August 19,2001 by Henikoff, Ahmad 
and Malik) that mentioned centromere variability as a mechanism for 
speciation. Perhaps this is an example of species selection?

The lack of significant number of species reproducing asexually has 
sometimes been cited as an example of species level selection - the 
asexually reproducing species simply don't last very long.


> A head count of the resulting species would not work as well as it
> does with individuals - since species can vary in size dramatically.
 
>> 3. What species-level entity takes the role that genes and traits
>> play in individual-level selection?  It must be a set of
>> species-level traits. Furthermore, they have to be traits that are
>> "emergent" at the species (or at least population) level.  They
>> cannot be "colligative" traits - for example, they cannot be
>> frequencies of individual genes, nor can they be averages of
>> individual traits.  They cannot, in fact, be quantitative features at
>> all - they probably need to be discrete, qualitative traits. (I
>> believe that this viewpoint has been championed by Elizabeth Vrba.) 
> 
> [...]
> 
> This is all after distinguishing "species selection" from
"species 
> sorting" - and *only* considering the former.
> 
> Species sorting may be important.  It may produce traits in
> individuals that apparently make little sense in terms of short-term
> selection pressures - and /only/ make sense when considering the
> selection pressures present during occasional periodic extinctions.
> 
> Your list of "species selection" qualifiers sounds very restritive to
> me.  


> It doesn't take much to qualify as an emergent trait.  For instance, 
> anything that affects the resulting speciation rate would immediately 
> qualify, in this context.

Right. These properties are going to be "emergent", since they will only 
be important at the species level. They have to be traits that differ 
between species, but this is true of many traits. I suppose "average" 
traits would be tougher to work with at the species level, but I think we 
need to list some examples. For instance, size might be considered an 
"average" trait, but a species with small individuals occupying a similar 
ecological niche will have a larger population than a species with larger 
individuals, and will be more likely to survive an extinction event. 

Yours,

Bill Morse
---
þ RIMEGate(tm)/RGXPost V1.14 at BBSWORLD * Info{at}bbsworld.com

---
 * RIMEGate(tm)V10.2áÿ* RelayNet(tm) NNTP Gateway * MoonDog BBS
 * RgateImp.MoonDog.BBS at 3/11/04 8:35:07 PM