Example of De Novo Evolution of a Gene
You know that natural selection we are talking about. That’s the natural selection that can not evolve a gene from the beginning. So your perceptual systems have failed you with the theory of evolution, this has been shown to you mathematically using the ev computer model
There is a great article on flower evolution in the June 2006
Natural History (pp 34
Origins of Floral Diversity) that quite clearly explains how a gene can evolve de novo: something you claim can't happen except by divine intervention.
It happens by duplication of an existing gene and then subsequent mutation until the duplicate endows an advantage to the organism and becomes a new gene.
I'll create an illustrative example:
We begin with a successfully reproducing flower. In a genetic accident, a gene becomes duplicated in the next generation (known to happen). This may be inconsequential at first, but when one of the duplicates mutations (e.g. by point mutation), one of four things can occur:
1) It does something harmful, and that plant fails to reproduce;
2) It makes no difference and the plant reproduces because it still has a good copy of the gene;
3) The modified duplicate causes a neutral change in the plant;
4) The modified duplicate causes an advantageous reproductive change.
#2 can happen repeatedly until #1, #3, or #4 can happen. When #1 happens there is little consequence: one plant, out of perhaps millions like it, just doesn't reproduce.
However, #2 and #3 can occur repeatedly without negative consequence until #4 occurs: an advantageous mutation to what then becomes a
unique new gene.
Suppose that new gene causes each of the flower's pedals to split in two right down the middle. Our flower species already has a symbiotic bee which finds these pedals quite attractive. Now the bee suddenly sees a flower with twice the number of pedals, and it is doubly attracted to it over the unmutated flowers. In very little time, the bees vigorously pollinate the mutated flowers, and the flower's unmutated bretheren die off. They may die off so thoroughly that the origin of the new gene may be completely obscured as the new gene is perfected through further mutations.
This scenario could easily be applied to the flower's pigment composition, pattern, pedal shape, size, etc.
The problem with applying Ev to de novo gene creation is one that, I believe, all computer models suffer. We can model something with a number of pre-programmed "knobs" to adjust, say, a flower's pedal count, size, spacing, or intensity of pigmentation. However, a gene evolving de novo may require, in a simulation, the spontaneous creation of a "new knob." The last time I studied computer simulations, we knew of no way to implement a knob that, when turned, makes novel knobs form. It is easily shown that evolution, as in my illustrative example, comes up with new knobs.
This is just one instance, and I think a critical one, of how your use of Ev cannot prove what you think it proves: that evolution is mathematically impossible.
Dr. Kleinman, I await your refutation.