inkorrekt writes:
Mutation and selection cannot produce new information.
As the other replies have made clear, you haven't provided a useful definition of information. A useful definition would provide a method for quantifying information in order to test assertions, such as that mutation and selection cannot produce new information (Gitt's Theorem 9 verbatim).
Using Shannon information (again, his landmark paper can be found at
http:///DataDropsite/Shannon.pdf) we can measure how much information is present in a population for a given gene. For example, the gene for eye color in a population of squirrels may include three different colors: red, green and brown (the different types of a gene are called alleles, so therefore this gene for eye color has three alleles: red, green and brown). Three eye colors can be represented in log
23 = 1.59 bits of binary data. The number of bits of information needed to represent three eye colors can be used as a measure of the amount of information contained in the eye color gene for this squirrel population.
Now let's say that one squirrel couple gives birth to a baby squirrel with a mutation in the eye color gene that gives its eyes the color yellow. The population of squirrels now has four eye color alleles for the eye color gene: red, green, brown and yellow. The amount of information needed to represent four alleles of eye color is log
24 = 2 bits. So the amount of information in the eye color gene has increased from 1.59 bits to 2 bits, an increase of .41 bits.
The new baby squirrel grows to adulthood and has many children, some of whom inherit the yellow eye color allele. The new allele gradually spreads through the population. Since the new eye color has not been selected against, it becomes a new allele of the eye color gene for the squirrel population, and the amount of information contained in that gene has increased by .41 bits.
This falsifies Gitt's Theorem 9 that mutation and selection cannot produce new information.
--Percy