john_SPAM{at}wilkins.id.au (John Wilkins) wrote in
news:c82urc$2v9h$1{at}darwin.ediacara.org: 

> William Morse  wrote:
 
>> You seem to be using this argument as a counterexample to the
>> possibility of species selection.  I have elsewhere argued that
>> traits that increase the overall carrying capacity of the environment
>> for the organism in question are an example of selection "for the
>> good of the species" (actually I would only say for the good of the
>> deme). I believe I can show how these traits will be favored even if
>> they have a small cost for the individual .
 
> Like Perplexed Jim, I'm losing track. I think all I want to say is
> that merely inheriting a range doesn't count as niche inheritance. The
> organisms have to make the niche, or improve it and redecorate a bit.
 
> The pigmy versus large heffalumps example merely goes to show that
> organisms cannot always modify their environment, and must sometimes
> adapt to it rather than vice versa, which I am sure is massively
> unsurprising a realisation.
 
> I'd love to see your argument for group good selection (notice how
> carefully I avoid species versus deme versus kin group distinction
> there?)...

Let me first note that the argument is not primarily for group selection 
as opposed to individual selection, but is simply that there can be 
traits that benefit the group as well as the individual, and these traits 
will tend to be selected via inclusive fitness. 

Any trait that increases the fitness of an individual without increasing 
the carrying capacity will necessarily decrease the fitness of all other 
members of the species in the affected niche (you gave the term for this 
type of population in another follow but I don't recall it). If a deer  
becomes fleeter than the other members of the herd, that will not allow 
more deer to survive, it will only guarantee that the fleeter deer will 
survive instead of a less fleet deer.  

On the other hand, a trait that increases the carrying capacity will  
increase the fitness of the other members of the niche-deme. Even though 
the trait is not directed at kin (e.g. elephants knocking down trees), 
since the members of the population that receive the most benefit will be 
those that are geographically closest to the individual, and will most 
likely be more related, the trait will spread even if it slightly 
decreases the individual's fitness. 

As you have probably noted,  my argument depends on  a degree of kin 
selection.  Members of the affected  group ( all those individuals which 
share the ecosystem that can be affected by the actions of another 
individual) must be more closely related to the individual than the 
species average.  A mutation that caused a salmon to swim up a different 
stream than the one it was born in, and then die,  would not be selected 
for, even though it would still increase the carrying capacity of the 
stream for frylings (or whatever the heck young salmon are called). The 
only reason I am describing this effect as different from kin selection 
is that a trait that increases carrying capacity does not have to be 
directed specifically at kin, and may thus appear to be one that is "for 
the good of the group". An example is the toxicity of the monarch 
butterfly. IIRC,  examples such as the monarch butterfly were used by 
Hamilton, but more recently attention on "kin selection" has seemed to 
focus more on actions that specifically benefit kin. And Hamilton's 
formulation places the emphasis on benefit to kin, rather than change in 
carrying capacity. 

Note that because the benefit per individual will generally be small (new 
carrying capacity less old carrying capacity divided by the affected 
population), the cost (if any) to the individual must also be small.  But 
by the same token, because such traits represent positive sum games with 
regard to the species, they have signficant advantages over traits that 
only benefit the individual. 


Yours,

Bill Morse


I can express this argument in mathematical form, but it is awkward on a 
newsreader.


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 5/19/04 6:42:52 AM