[continued from previous message]   
      
   > Stabilising selection around workable compromises   
   >   
   > So when you observe “thousands of traits conserved”, what you are seeing   
   > is:   
   >   
   > populations remaining inside a feasible region of phenotype space.   
   >   
   > 7. Why selection doesn’t “collapse under dimensionality”   
   >   
   > Putting it together:   
   >   
   > Selection avoids a combinatorial explosion because:   
   >   
   > Most dimensions are neutral or nearly neutral   
   >   
   > Purifying selection removes large disruptions efficiently   
   >   
   > Pleiotropy couples traits, reducing effective dimensionality   
   >   
   > Selection acts on organisms, not traits   
   >   
   > Drift absorbs small-effect noise   
   >   
   > Stabilising selection maintains systems near viable equilibria   
   >   
   > This is why mammals can persist with:   
   >   
   > Highly conserved core systems (development, metabolism, cell biology)   
   >   
   > Slowly drifting peripheral traits   
   >   
   > Occasional adaptive shifts when environments change   
   >   
   > 8. A useful analogy (non-hand-wavy)   
   >   
   > Think of a modern aircraft:   
   >   
   > Thousands of parameters   
   >   
   > Most cannot change without catastrophic failure   
   >   
   > Some can vary within tolerances   
   >   
   > Rare redesigns improve performance under new constraints   
   >   
   > Natural selection is not “designing” the aircraft from scratch each   
   > generation; it is:   
   >   
   > Rejecting planes that crash   
   >   
   > Allowing minor tolerated deviations   
   >   
   > Occasionally locking in improvements when conditions demand it   
   >   
   > 9. Where this intersects ID-style intuitions (briefly, since you follow   
   > the literature)   
   >   
   > The intuition that “too many traits must be coordinated” is not wrong,   
   > but the resolution is:   
   >   
   > Coordination is largely inherited, not newly created   
   >   
   > Selection is primarily conservative, not creative   
   >   
   > Evolution is maintenance with occasional modification, not constant   
   > innovation   
   >   
   > This is precisely why:   
   >   
   > Evolution is slow   
   >   
   > Major innovations are rare   
   >   
   > Lineages show long periods of stasis punctuated by change   
   >   
   > Summary   
   >   
   > Thousands of traits do not need to be optimised simultaneously.   
   > They are:   
   >   
   > Mostly conserved by purifying selection   
   >   
   > Buffered by developmental and genetic architecture   
   >   
   > Allowed to drift within tolerances   
   >   
   > Occasionally modified when local fitness gradients align   
   >   
   > Population genetics handles this cleanly once you stop thinking in terms   
   > of independent trait optimisation and start thinking in terms of high-   
   > dimensional, constrained, organism-level selection.   
   >   
   > If you want, next we can:   
   >   
   > Walk through a toy population-genetic model with many traits   
   >   
   > Discuss genetic load and Haldane’s dilemma in this context   
   >   
   > Or connect this explicitly to Evo-Devo constraints in mammals   
   >   
      
   What did you learn from this?   
      
   ID is still a bait and switch scam because selection was never what you   
   thought that it was.  How do you think that Behe's three neutral   
   mutations as evidence for design works?  He has to put a time limit   
   because when you deal with reality you find out things like mice and   
   humans have been evolving as different lineages for around 80 million   
   years.  Most of their proteins may be around 20% different, so for an   
   average protein of 300 amino acids there are around 60 neutral mutations   
   that have occurred between mice and humans at this time.  These proteins   
   still do the same things in both mice and humans.  We can take human   
   proteins and put them into mice and they work.  If some mutation occurs   
   that changes the function of that gene a lot of those neutral mutations   
   may be involved in the new function.  They might not have changed the   
   gene function enough to be selected for or against, but they could   
   combine with other mutations to do something different.   
      
   Behe needs his time limit because of the way that protein genes evolve.   
   Your new information is forming all the time, and is how new traits   
   evolve and are selected for or against.   
      
   Everything has to work with what is already working.  Whales lost a lot   
   of genes because they didn't need them anymore, or they did things that   
   were counter productive in the new environment that the whales were   
   existing in.   
      
   Ron Okimoto   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)