Subject:  Re: Pigs could flya - Reply         
Date: Wed, 29 May 1996 10:02:55 -0400 (EDT)



> And Paul, what determines a species "longevity" ?  Does not the               
> persistence of a species require the survival and reproduction of many        
> of its [individual] members?  I'd like to be clear on exactly what the        
> issue is, expecially if I'm to be indicted.                                   
>                                                                               
> Thanks,                                                                       
> Lisa                                                                          
 
Well, that's a tricky question.  Species selection results from those
agents that affect extinction and speciation.  You then argue that anything
that results in different rates of reproduction and survival (natural 
selection) affects extinction and speciation in some way.  I suppose the 
answer I'd give is that some agents are much more important than others
in causing extinction.  As a result, the separation of the concepts of
natural selection and the special case of species selection is useful.
Also, natural selection can result in the organism changing to affect
its reproduction and survival.  Species selection doesn't do this, in part 
since most change is concentrated in speciation events and species remain
relatively unchanged afterwards.  

I'm not sure of my answer.  Certainly it could be expressed better.  
I went to the library and looked at Stanley's Macroevolution,  I'll
try to summarize some of his arguments.                   
                                                                                
Stanley argues that large-scale evolutionary change is dominated                
by species selection.  Species selection is the result of differential          
rates of speciation and extinction among species.                               
                                                                                
Now, the actual process of speciation might be guided entirely by               
natural selection.  But since speciation events occur predominantly             
in small populations, the direction of change occuring at the time              
of speciation will be directed largely by "historical accident -              
in particular, by the nature of the habitat in which a potential                
ancestral species happens to exist, by the availability of niche                
space for potential occupancy, and by the nature of geographic changes          
that happen to effect divergence.(186)"   Also, given several speciation
events, some new species will move into one niche, some into another.
And species encounter a variety of different habitats, each with
different selection pressures and available niches.  So, speciation          
will have a large random element even if evolution is guided solely             
by natural selection.                                                  

The punctuational model of speciation emphasizes how "opportunistic"
and "experimental" evolutionary change is (p. 212).  Speciation occurs
relatively quickly in response to local conditions.  Whether these
changes will benefit the organism in a wider environment is uncertain. 
Whether the changes brought about by several speciation events will
add up to an adaptive breakthrough seems like a matter of chance.
(My comment: hence, a weakness of adaptationism: evolutionary change
is primarily a matter of adaptation to local contingent circumstances,
even if it is guided entirely by natural selection.  Change is not 
usually, and not necessarily, a response to the circumstances faced by
the organism in a larger environment.)
                                                                                
Once species form, Stanley argues, species selection dominates.                 
"If rapidly divergent speciation interposes discontinuities between
rather stable entities (lineages) and if there is a strong random
element in the origin of these discontinuities (in speciation),
then phylogentetic trends are essentially decoupled from phyletic
trends within lineages (p. 187)."

Species selection is not group selection as described by Wynne-                 
Edwards.  Wynne-Edwards suggested that selection could exist for                
traits that that are relatively harmful for individuals but valuable            
for semi-isolated populations.  Species selection does not result in            
adaptations becoming fixed within species because they are valuable             
to the species, rather than to the individual.  Adaptation results              
primarily from selection acting at the level of the individual.                 
What species selection does is determine the "fate of adaptations,
onced established."

The agents of species selection are "the factors that limit population
size and cause extinction: competition, predation (including disease),
habitat alteration, and random fluctuation in population size(p. 211)."
That is, the agents of species selection affect population density.
The same factors that contribute to extinction contribute to speciation,
because "failure to speciate represents a failure of small population
to expand into a new species.  When conditions favoring a successful
taxon deteriorate, we must expect not only that its lineages will go 
extinct at a higher rate, but also that remaining lineages will be less 
likely to cast off isolates that blossom into new species. (pp. 197-198)"
Stanley thinks biotic change and climatic change are more important than
other forms of changes in the physical environment, since they tend to
change in a directional manner rather than fluctuating randomly.  Also,
he thinks competition has been overemphasized as an agent.  He thinks
predation is more important for "behaviorally simple, heavily-preyed
upon taxa."

Finally, I'll mention an example Stanley uses to show how each speciation
event can move in some adaptive direction while the overall direction of
divergent speciation remains random: "a heuristic ananology - though one
representing a system that is more perfectly random - would be a coin-tossing
exercise in which some coins are unweighted, but of the rest, half are
weighted in favor of heads and the other half are similarly weighted in
favor of tails.  If the identity of the next coin to be flipped is
determined by chance, heads and tails have the same probability of
appearing (p. 187)."


Paul
pcg-AT-panix.com



     --- from list marxism2-AT-lists.village.virginia.edu ---