I'm genuinely mystified as to what Dennis' problem is.
Surely he must concede that e.g. point mutations *exist* and are random (which can be demonstrated). Then it is common sense that sufficient pile-up of random mutations is capable of transforming any sequence of DNA into any other conceivable sequence, which would include functional genes.
Here's a scenario: at the start a mutation duplicates a functional gene. The offspring is viable. Now random mutations start to strike. If one strikes the first copy and cripples it, and another strikes the second copy and cripples it: dead, no offspring, end of line. If one strikes the first copy and cripples it, but none strike the second (and vice-versa), or the mutation turns out neutral, then there are no consequences for procreation but a first step is taken in the transformation of a gene to a possible other functional.
Considering the inconceivable amount of offspring that a bacteria can produce in just days or weeks, there will always be a sufficient amount of viable varieties coming out of the filter of natural selection (with a spare duplicate dysfunctional gene waiting there as target for yet more mutations). In the end millions or billions of permutations will be tested in the copy of the gene that became crippled without consequences. At some point one can expect that something functional arises. And even if not, there are no consequences either. At any moment the genome can contain tens of thousands of such "laboratories" that do nothing - either positive or negative - for the time being.
From a distance, the net result of these mechanisms looks a lot like directed design. But that's only because the "backoffice" activity remains completely hidden unless specifically monitored.
The key to grasping this, is understanding that a typical living being happens to be... DEAD. What we see, the survivors, is only the top of a giant iceberg of failed natural experiments. It's a numbers game in which all but a minority fail. But the failures remain largely hidden since they don't live, and since the living are more than sufficient to quickly fill the entire available niche anyway (calculate the evolution of an unimpeded elephant population to get a nice illustration: in just a couple of thousands of years they will cover the entire solar system with elephant dung).
Design is produced by randomly shotgunning mutations in combination with huge numbers of trails, and there's nothing magic about it.
quote:I'm genuinely mystified as to what Dennis' problem is.
Mine is with evolution Anna.
Obviously. And I see how it works: you have decided that you're going to have a problem with it, no matter what. You will accept no other outcome to your breakdown of the fundamental mechanisms of random mutation and natural selection, than the conclusion that it must be impossible. And in order to reach that outcome, you have no other choice but fostering your "confusion", juggling meaningless concepts as "genetic loss" and "information", and declaring unequivocal experimental evidence as "doubtful"...
quote: If you read back on this thread, you will find some debatable examples of genetic mutational gain (E. Coli), and on another thread, about Nylonese bacteria (both interesting reads). Though even if these arguable examples were true,
See what I mean? (bolded) There's nothing debatable or questionable about these results, but you NEED them to be debatable or questionable to maintain your position, so you do what needs to be done.
Excuse me if I skip the part where it would take entire books to respond to all the confusion...
quote:From a distance, the net result of these mechanisms looks a lot like directed design.
Because it is. Though this is yet to be determined, since I haven't even gotten to supporting my viewpoints yet.
quote:giant iceberg of failed natural experiments.
I would ask for your evidence, but I suppose you don't have any, since there are no references to support anything said here.
Evidence, although plenty available, isn't even needed. Math is enough as you will see below.
quote:It's a numbers game in which all but a minority fail.
Evolution is a game?
quote:calculate the evolution of an unimpeded elephant population to get a nice illustration: in just a couple of thousands of years they will cover the entire solar system with elephant dung
Hahahahaha. This (if it is true) is actually an excellent point for ID. If elephants were, at one point, the most advanced of land organisms, what predators would kill them? Another example of overpopulation due to evolution over millions of years is flies.
If two flies were left to reproduce over a year, without predators or other limitations, the resulting flies would cover the entire earth.
I'm not sure what your point was there with the elephant poop, but it was really funny, and really bad.
There's an applet that allows you to roughly calculate (unrestricted) elephant population numbers over time. If we make a couple of reasonable assumptions (6 infants per pair of elephants, fertile at 30, productive for 60 years), you get these kind of results from one pair of elephants:
After 200 years 54 elephants After 500 years 13100 elephants After 750 years 1.060.000 elephants After 850 years 9.600.000 elephants After 900 years 28.700.000 elephants After 1000 years 86.000.000 elephants After 1030 years 258.000.000 elephants ... After 2000 years 11.100.000.000.000.000 elephants
Obviously you can nitpick on the assumptions and exact formula, but the numbers are in the right ballpark at least. Elephants have existed for millions of years, so even if we would have to multiply the timescales by 10 or even 100 to be closer to real, there's more than enough time to easily surpass these numbers.
Now, the highest stable elephant population, before man started to hunt them extensively, was in the order of 10.000.000. So obviously, something is going on! Also remember that elephants have no natural enemies worth mentioning (if we forget about untypical mankind with its current resources for a moment).
Of course all this is no mystery: the population is mainly limited by the available resources (water, food, territory etc.). The earth is simply not able to support more than roughly 10.000.000 elephants. But you have to let sink in what this means: since the elephants' reproduction rate is largely independant of the state of the population, it means that even when that population is at its 10.000.000 maximum, it is at that very moment in the process of producing 18.000.000 new elephants over the next 50 years! Even at the moment when the actual ceiling is already reached. Which means that roughly twice as many elephants will die over that 50 year period, as the ones that remain alive. This process repeats itself over EVERY next 50 year period. For one of the SLOWEST reproducing animals. It doesn't take much imagination to see what would happen if you do this exercise with mice, flies, let alone bacteria. This is what I meant when I said that the typical state of a living being is "being dead". The living ones are the -much more visible and therefore misleading - exceptions.
The underappreciated consequence of all this is, that negative mutations are irrelevant. Since selection (the competition among the elephants for available resources) is non-random, they invariably disappear into the reservoir of elephants that die anyway. The reservoir of twice as many elephants as living ones that is by necessity eliminated every 50 years, EVEN if we would suppose that there would be exclusively positive mutations for some miraculous reason. Negative mutations are merely background noise, a necessary by-product of the mechanism. Positive mutations, no matter how rare, are on the other hand virtually guaranteed to be preserved by the advantages they bring to their carrier.
And this is why randomly shotgunning mutations at the genome ends up over time with the same result as from time to time carefully inserting a rare positive one.
If you want to make an argument that any mutation at all to a protein is going to result in diminished function, you can try, but that argument is certainly wrong; there's an enormous amount of evidence that mutation rarely alters protein function at all, and many proteins have not been under enormous selection pressure to optimize, so there's much room for mutational improvement.
I think this is an example of an extremely important point here, which is generally overlooked, and the ultimate source of a lot of disbelief in the workability of the mechanism of mutation and selection: the make-up and workings of DNA and proteins are very, very messy. This is not somekind of extremely finely tuned, highly optimized mechanism that is inevitably degraded significantly by random mutations. Rather, in their very processes they have a lot of tolerance built-in. The secret to the processes of life is not perfection, but flexibility. The trade-off is that there is a certain degree of sub-optimality: life is rather "good enough" than "perfectly tuned".
One needs to appreciate this before being able to grasp how a constant bombardment of random mutations does not necessarilly immediately cripple the whole thing, and *can* be the basis of constructive effects in the longer term.
But this seems to be a very hard concept to grasp for people who have problems with continuum (non-discrete) thinking in general, and who have difficulty appreciating the fuzzynes of species, of life/non-life, of personhood/non-personhood, of good/bad...