Perhaps part of the confusion is due to a lack of definition: what exactly would constitute a reversal of evolution? IOW, do you even know what you're asking?
RAZD writes:
That the vein patterns in the wings are the same means that the vein pattern is not lost, ... but then this is the vein pattern in the part when it is NOT a wing as well as when it IS a wing. But the question remains: if the wing is turned OFF by one mutation, and then later is turned ON, then is this a new mutation of a new ON switch, or is this a new mutation that reverses the mutation that turns the wing OFF? It is entirely possible that both could occur.
I think it's more probable that the information is not lost and hence in this case evolution can reverse something that was done millions of years ago.
Did you understand waht RAZD was describing? Do you have any idea where that information resides? And just how is the retention of that information supposed to indicate evolutionary reversal? The conclusion you reached seems contrary to the premises.
Are you familiar with basic genetics? Familiar with what the genotype and phenotype are and how they related? Furthermore, do you
understand about genotypes and phenotypes and how they fit into evolutionary theory? If you do not, then no wonder you're confused.
The phenotype is that set of characteristics that the physical body has. It gets those characteristics from the genotype, which is that individual's set of genes. It is the genotype that contains the genetic information for the
expression of physical characteristics in the phenotype. Note, please, that said information
may or may not be expressed; there are specific DNA sequences that will switch genes on or off.
Thus, the information can exist in the genotype, but never get expressed in the phenotype. Like hens' teeth. Do chickens have teeth? No, but the dinosaur-bird transition, represented by Archaeopteryx, did. Bird "lost" their teeth ... or did they? In experiments in which mouse-embryo gum tissue is placed onto chick embryo jaws, those jaws develop tooth buds. Thus we see that chickens' genotypes still contain the genes for teeth, but those genes are normally never expressed and thus never appear in their phenotypes.
Now, how do genotypes and phenotypes fit in evolutionary theory? What the phenotype is like depends directly on the expression of the genotype. All variability lies in the genotype -- while that variability can result in variation among the population's phenotypes, it is still the variations within the genotypes that is of significance, which is one reason why we call it
genetic variation. All mutations that have any evolutionary significance occur solely in the genotype and
specifically in the genes of the germ cells (ie, sperm, eggs) and not in the body cells; only those genetic changes that can be inherited could ever be of any evolutionary interest. All recombination through reproduction also happens solely on the genetic level, in constructing a new genotype.
The only real role the phenotype has is that selection doesn't act upon the genotype directly, but rather indirectly through the phenotype. The different phenotypes get selected for or against and so indirectly their genotypes get selected. It's a very results-oriented process.
So, how do we measure whether evolution has been reversed? Just by looking at the phenotype? No, because come traits can be switched on or off with minimal changes in the genotype. I would vote for the genotype, since that is where the information actually lies. For evolution to reverse, all the genetic information for the trait in question would need to have been removed from the genotype. As I understand, that very rarely happens; rather, those genes are either switched off or damaged a mutation or by a ERV -- eg, we and other apes cannot produce our own vitamin C because that gene was destroyed by the same identical ERV, so we need to get it from our diet.
Another scenario is that a different genotypical pathway gets followed to "reverse" evolution -- eg, cetaceans having evolved from land mammals to become aquatic and "fish-like". But whereas it may superficially appear that evolution has been reversed, in reality a lot of new changes have been made to the genotype. The true test of whether evolution has been reversed is to compare the genotypes: only if the genotype has reverted to the earlier form can we even begin to suggest that evolution has been reversed.
Just in case you couldn't follow that, let me offer an analogy. Progressive dances proceed around the floor counter-clockwise along the "line of dance" (LOD). You enter the floor with your partner and start to dance down the LOD. A number of times you will pass by that point at which you had started. Now, have you reversed your dance? Sure, there are a few points where you had momentarily moved backwards, but you can ill afford to do that too much nor for too great a distance for risk of the other dancers crashing into you (country two-step is a particularly good example if this, since even slowing down can cause you be be run over and, let's face it, red-necks don't brake for nobody). Reversing your dance would require you to proceed
against the LOD (ALOD). But didn't you instead proceed and progress ever
forward in order to arrive at your starting point?