Guy Hoelzer  wrote in
news:bmfbp4$1g5t$1{at}darwin.ediacara.org: 
>> A pack of wolves is not a reproducing agent as far as the pack is
>> concerned - it is composed of individual wolves that choose to join
>> or stay in the pack. But invidual wolves without a pack will not
>> survive long, so their heritable fitness differences depend on the
>> fitness of the pack. 
> 
> Well, there is quite a large literature on the reproduction of social
> groups.  I am not sure of the story with wolf packs, but I would be
> surprised if they did not reproduce like similar kinds of social
> groups in other species.  I have done a fair amount of population
> genetic and phylogenetic work on macaques, in which social group
> reproduction (splitting akin to cellular division) is well documented.
>  It doesn't matter at all in my view that individuals can choose to
> join or leaves groups.  Also, I am not comfortable with what you are
> calling "the fitness of the pack."  You seem to be referring to the
> quality and/or quantity of resources provided to individual wolves by
> virtue of pack membership, which does not bare any resemblance to the
> notion of evolutionary fitness. 

I have had the impression throughout this thread, as you have noted, that 
our disagreements are partly semantic. Of course, often semantics can 
mask a fundamental disagreement. That is one reason why I like examples. 
And in this case the most intriguing example to me was the social group, 
so I would like to focus on that, and start a new thread. 

I had envisioned the group (wolf pack, macaque ?troop?) as having to 
reproduce  through individuals because otherwise everyone in the group 
would have to split in two. So what I was really talking about was 
replacement of the group (which is one form of reproduction) by 
reproduction of individuals. You are thinking, at the group level of 
selection,  of a group growing by accumulation of individuals and then 
splitting, without creating a "copy" of the old group. Interesting. My 
objection of requiring a replicating agent at a lower level of selection 
might be moot in this case, although on the other hand there might be so 
little reproductive fidelity (i.e. no "heritable variation") that 
selection may not operate. I suppose I would have to actually read some 
of the "large literature"- and it is so much easier to just spout off in 
my "dangerous" mode (the one based on a little knowledge). 

As to my reference to "fitness of the pack" - without doing some math, I 
won't guarantee that my reference isn't just fuzzy thinking. But if some 
packs fare better than other packs, if more packs are created than can 
survive, and if whatever causes the different success of packs is 
heritable _by the pack_, then you should get group selection of the pack, 
and the pack should have "fitness".  The third condition is the kicker. 
Let us assume that packs do well based on a combination of cohesiveness 
and a good leader. Kinship can take care of some of the requirements for 
group selection in this case, but what if there is also an element of 
individual tendency to recognize and support the best leader within a 
pack, to join a pack with a good leader and abandon a pack with a poor 
leader, and to ostracize non-cooperative pack members? Packs composed of 
such members will on average do better than other packs, which will in 
turn enhance the reproductive success of the pack members, which will in 
turn create more "good" pack members that will reproduce better packs. 
But the traits that are useful in a pack will be useless in the absence 
of a pack. So is this not group selection at the level of the pack?

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 10/23/03 6:15:10 AM