Misschien zal de uiteenzetting van Mark Ridley je iets meer op het spoor zetten:
(en ja het gaat weer over de motten

)
--------------------------------------------------------------
4.6.2 Mutation is random with respect to the direction of adaptation
When the environment changes, as it did for British peppered moths in the nineteenth century, a new form of the species comes to be favored. According to Darwin’s theory, natural selection will favor whatever form, among those represented in the species, fits the new environment best. The environmental change does not itself cause mutations of the right form to appear. New mutations of all sorts will be constantly arising, independently of what is required for adaptation to the current environment; selection works on this variation. Such is the Darwinian theory. The alternative would be some kind of directed mutation. If mutation is directed, then when the environment changed to favor melanic moths, melanic mutations would be generated by some mutational mechanism.
There are both factual and theoretical reasons to doubt that mutations are adaptively directed. The factual evidence comes from laboratory observations of spontaneous mutations. There is no evidence that these mutations take place toward the adaptive needs of the organism. Until recently, a classic experiment of Luria and Delbruck in 1943 was the “textbook” demonstration that mutation is undirected; but that particular experiment was effectively challenged by Cairns et al. in 1988 and is currently the subject of active research.
The strongest reason to doubt that mutations are adaptively directed is theoretical. In the case of the peppered moths, melanic moths probably already existed in the population at the time of the industrial revolution, and selection %imply increased their frequency; but we can use the case as an example as if there were no melanic moths when the environment changed. The industrial revolution imposed on the moths an environment they had never encountered before. The environment (probably) was completely new. If there were no melanic moths in the population, a mutation, of melanic coloration, was then needed for the moths to survive. Could it arise by “directed” mutation? At the DNA level, the mutation would have consisted of a set of particular changes in the base sequence of a gene. No genetic mechanism has been discovered that could direct the right base changes to happen.
If we reflect on the kind of mechanism that would be needed, it becomes clear that an adaptively directed mutation would be practically impossible. The organism would have to recognize that the environment had changed, work out what change was needed to adapt to the new conditions, and then cause the correct base changes in the relevant parts of its DNA. It would have to do so for an environment the species had never previously experienced. It would be like humans describing subject matter they had never encountered before in a language they did not understand: like a seventeenth century American using Egyptian hieroglyphics to describe how to change a computer program. (Hieroglyphics were not deciphered until the discovery of the Rosetta Stone in 1799.) Even if it were just possible to imagine, as an extreme theoretical possibility, directed mutations in the case of moth coloration, the changes in the evolution of a more complex organ (like the brain, or circulatory system, or eye) would require a virtual miracle. It is for this reason that mutation is thought not to be directed toward adaptation.
Although mutation is random and undirected with respect to the direction of improved adaptation, that does not exclude the possibility that mutations are non-random at the molecular level. The higher frequency of transitions than transversions is a case in point. It is not true that changes from the base A, for example, are equally probable to C, I, or C; a change to I may be 10 times more probable. Other molecular biases are also known. One example is the tendency of the two base sequence CC to mutate, when it has been methylated, to TG. (The DNA in a cell is sometimes methylated, for reasons that do not matter here.) After replication, a complementary pair of CG on the one strand and CC on the other will then have produced TG and AC. Species with high amounts of DNA methylation have (probably for this reason) low amounts of CG in their DNA.
These molecular mutational biases are not the same as changes toward improved adaptation, however. You cannot change a light-colored moth into a dark one just by causing transitional, rather than transversional, mutations in its DNA, or by converting a proportion of its CC dinucleotides into TC. Some critics of Darwinism have read that the Darwinian theory describes mutation as “random,” and have then trotted out these sorts of molecular mutational biases as if they contradicted Darwinian theory. But mutation can be non-random at the molecular level without contradicting the Darwinian theory. What Darwinism rules out is mutation directed toward new adaptation. Because of this confusion about the word “random,” it is often better to describe mutation not as “random,” but as “undirected” or “accidental” (which was the word Darwin used).
One last point should be made about assessing evidence for directed mutations in laboratory mutations. The mutations studied in the laboratory, as we saw, are likely to be a biased sample. They are likely to be the more frequent kinds of changes. Now, it is possible that some of these frequent changes are not spontaneous point (or chromosomal) mutations; some may be caused by the insertion of mobile genetic elements (section 10.6, p. 255). For some purposes, this does not matter. A spontaneous genetic insertion can be just as much a mutation as a change in a base or a chromosome. However, insertion elements may in some cases have an organized, probably adaptive, genetic structure; they may therefore be able to confer whole new adaptations on the organism into which they insert, like plasmids in bacteria. If the organism could in turn influence the chance that insertion elements will be mobilized inside itself, then a sort of directed mutation might appear to be operating. It is thus important, when assessing evidence for directionality in mutation, to know what the genetic basis of the alleged mutation is. The Darwinian claim that mutation is not adaptively directed applies only to spontaneous new mutations, not to the activation of previously adapted genetic sequences. This is a hypothetical possibility; but the claim that mutation is undirected is so fundamental to Darwinism that we must understand exactly what it means.
-------------------------------------------------------------