This is one of the sillier things you've asserted. Genes don't fade away, and a single mutation that a sexually reproducing creature has will appear in half his offspring. If the mutation trait is dominant, and beneficial, it will almost certainly maintain itself and take over the population in short order. If the mutation trait is recessive, it will be entirely hidden in the second generation, and will rarely appear in subsequent generations, and its growth or elimination is more a matter of chance, but it will not rapidly disappear.
To take an example that works just like the former mutation case, if a person with a single gene for brown eyes (dominant trait) moves to where everyone has blue eyes, half of his offspring will have the gene for brown eyes, and of those, half of theirs. (The absolute numbers don't decrease though, because the number of descendants also doubles in each generation)
And being dominant, the gene will express itself visibly (brown eyes). Now, if brown eyes are a neutral trait, the gene is as likely to increase in count as to decrease, but since the number of people with it is not much larger than zero, with random drift it is more likely to disappear at some point.
But if brown eyes are beneficial, or for whatever reason the people with brown eyes produce more offspring and so on than people with blue eyes, then eventually the whole country will have brown eyes.
New mutuations, if they are neutral, are no more or less susceptible to disappearing than brown eye genes in this case, if they are neutral.
You need to take this up with the folks at Talk-Origins. You know the Darwinian bible folks.
From their Introduction to Evolutionary Biology
Neutral allelesMost neutral alleles are lost soon after they appear. The average time (in generations) until loss of a neutral allele is 2(Ne/N) ln(2N) where N is the effective population size (the number of individuals contributing to the next generation's gene pool) and N is the total population size. Only a small percentage of alleles fix. ...
Deleterious allelesDeleterious mutants are selected against but remain at low frequency in the gene pool. In diploids, a deleterious recessive mutant may increase in frequency due to drift. Selection cannot see it when it is masked by a dominant allele. Many disease causing alleles remain at low frequency for this reason. ...
Beneficial allelesMost new mutants are lost, even beneficial ones. Wright calculated that the probability of fixation of a beneficial allele is 2s. (This assumes a large population size, a small fitness benefit, and that heterozygotes have an intermediate fitness. A benefit of 2s yields an overall rate of evolution: k=4Nvs where v is the mutation rate to beneficial alleles) An allele that conferred a one percent increase in fitness only has a two percent chance of fixing. The probability of fixation of beneficial type of mutant is boosted by recurrent mutation. The beneficial mutant may be lost several times, but eventually it will arise and stick in a population. (Recall that even deleterious mutants recur in a population.) |
You are giving a lot of ifs in there. First is has to be dominant, that's a 50/50 chance. Then it has to be beneficial, the odds of that is the big problem. As I explained, neutral mutations will almost certainly die off. This is a mathematical fact proceeding directly from Mendelian genetics. It is irrefutable. Even slightly beneficial mutations will die off and not take over because the odds against a single mutation taking over a whole population. So all that could take over a whole population is a largely beneficial mutation. Problem with that is for any new feature, for any new function, indeed for any new gene you would need a multiplicity of mutations. These are not all going to occur at once. Since neutral and slightly beneficial mutations will die off very quickly and only spread to very few individuals, this accumulation of mutations is impossible and therefore evolution is impossible.