Evolution Simplified

It's Chiro's post, so I'll let him address the bulk of your misunderstandings, but I want to be a part of the discussion as well so I'll quickly address this one, myself.You're confusing what is meant by "predictable." Statistical observations about the odds are not predictions. A proper flip of an unweighted coin is unpredictable. The fact that you know that, in the past, it's landed heads about half the time and tails about half the time, doesn't mean you have any ability to predict the outcome of your next flip.You just don't know. It's unpredictable, even though you can say what the odds of it being one or the other are going to be. You still don't know the outcome in advance, so the flip is unpredictable.We can tell you the odds of one or another mutations happening, sure. But we can't tell you what mutation is going to happen next, for sure. We know the odds but the outcome cannot be predicted. Clear enough?

I don't understand why you're attempting to draw this dichotomy. The assertion was that mutations (new traits) were not predictable, not that evolution as a theory was incoherent and completely lacking in predictive power.Which is it? It's both. Evolution is used to make predictions; mutations happen at random and while they can be analyzed with statistical tools, the precise outcome and occurance of any single mutation cannot be predicted.If I may suggest - this is going to be a much more fruitful discussion for you if you concentrate more on learning the information that is being presented to you, and concentrate a lot less on trying to trip us up in amateur games of "gotcha."

But evolution isn't a system. I get that living individuals are systems, yes. And, indeed, living individuals are born and then they decay and die.But populations don't work that way, because they're not thermodynamic systems. They don't use energy; they're just groups of individuals. The group itself is not a thermodynamic system, so to say that the group evolves does not violate thermodynamics, because populations of living organisms are changing and advancing with a net increase of entropy in every individual over its lifetime.The fact is, even when old men father children, those children are not themselves geriatrics, but are young, fresh individuals. Every time an organism reproduces the "entropy clock" is reset. It is moot, because the history of living things on Earth is not a thermodynamic system, even though individual living things are. But since individuals don't evolve, evolution hardly violates the Second Law.

I don't get it. Why is that inconsistent with an old universe? It's not like hydrogen has a half-life, or anything.Is he talking about stars all burning it up? That doesn't seem like a reasonable criticism to me. There aren't enough stars. Well, the second one really isn't. Evolution is a theory of biology, not cosmology. So arguments from cosmology don't really apply.Have your brother show up here sometime.

We can show you all the evidence in the world, but until you understand the theory, you won't understand what the evidence actually proves. The evidence substantiates the modern theory of evolution, which combines Darwinian mechanisms of natural selection (and other forms of selection) with molecular genetics. The evidence doesn't substantiate:1) That evolution happens when individuals undergo spontaneous metamorphosis;
2) That evolution operates from some kind of pre-programmed master plan to violate the Second Law;
3) That evolution means we should all commit suicide to end our pointless lives;or other such nonsense.

We've observed that it has, and many of those observations have been presented to you. You've had no response in each case except to say "no, that didn't happen" and call your opponents liars, as you called me in another thread.It's starting to get a little ridiculous. You'll have to explain how that's something we'd expect to see. Humans giving birth to chimpanzees would disprove evolution, not be something we'd expect to see. Sequences are arbitrary. We can put any genetic sequence into any other, and the cell will produce those proteins.Because those proteins, though, are being produced in the wrong environment, we find that they often have detrimental effects. But often they don't. The fact that we can do it in the lab proves that it can happen in nature, too. In the natural world, the environment itself programs the policy. The environment determines what sequences are selected for or against, something that a designer does in genetic programming. The capability to what? I think you've failed to understand what genetic programming is.

You never played "rabbits and foxes"? It's pretty trivial to model the fact that organisms that need resources from the environment to survive reach an equilibrium with that environment, where their population size is balanced against the amount of the resource the environment can produce. We call the population size that the environment can support "K".

Well, I think he is, because he's trying to phrase these as observations. We observe that most populations have reached a plateau, and we have every reason to believe that those populations that haven't yet will. Yeah, pretty much.

If outcomes are statistically random, then every possible outcome occurs given sufficient time. (This is similar to the mathematical phenomenon that, given a statistically random series of numbers of sufficient length, like the decimal expansion of pi, you can prove that any arbitrary, finite sequence of numbers is contained within it.)Since there's no physical law apparently preventing alternate energy states and chemical outcomes of the molecular interactions involved in the replication of genetic material, we know that all those alternate outcomes are possible. And since they're possible, they'll happen given sufficient opportunities for them to do so, at random.So, yeah. Seems like, to me, there's a pretty good reason to assume that mutations are inevitable. The same reasoning applies to the question of those mutations being silent or not. Since there's a non-zero probability that a random change to a polypeptide sequence will alter the chemistry of the resulting protein, the outcome that such a change does alter that chemistry is inevitable, given sufficient time.Since it turns out that those things have a pretty substantial probability, that sufficient time turns out not to be very long.

That's a different question, though.As it stands, the rate for mutagenesis in mammalian nuclear DNA, in terms of single-point substitutions, is 1 per every 3.6 billion base pair replications, according to my wife's phylogenetics text.

Your genome is upwards of 3 billion base pairs long, and that genome is fully replicated every time one of your cells undergoes meiosis or mitosis. Which happens, in your body:1) hundreds of times every second in your skin as new skin cells and hair are grown and the lining of your digestive tract is replaced, plus
2) thousands of times per second as your bone marrow replicates and differentiates new cells to replace the contents of your blood; plus
3) 1.5 million times a day as your testes produce sperm (assuming you're a man), plus
4) whatever assorted other cell division is going on in your body.So, for every act of cell division per day that comes to (millions), we should expect the new sequence of DNA to differ from the old one in 1 or 2 different places.

If my post didn't answer your question, I guess I don't know what you mean by "frequent." We estimate that any one human being has between 5 and 500 mutations. Every single human. Would you characterize that as "frequent"? I'd say that it's "sufficient."

I would estimate that each pup would have between 5 and 500 mutations, individually. That's a wide range because I don't know how long the dog genome is, or the precise rate of mutation in nuclear dog DNA.Excellent way of framing the question, though. I'm sorry I wasn't clearer before. You're right, of course, that there's a difference between the mutations that are passed along from the parents' gametes and the somatic mutations that an organism naturally accrues in its lifetime. In a sense, an organism is born with all the mutations that it will ever possess that are relevant in an evolutionary context. It will then pass along those mutations, possibly; also, as it generates gametes, mutations will form that aren't relevant to the parent's fitness but become part of the genome of the offspring.This message has been edited by crashfrog, 05-12-2006 12:18 AM

Mutation includes changes that happen due to imperfect replication, as well as changes that happen when DNA is damaged and then repaired improperly. That's sexual recombination. Sexual organisms have two alleles per gene (usually), one of which they inherit from each parent. The phenotype of the offspring depends on how those two alleles interact. Those interactions are what people are talking about when they talk about dominant and recessive traits.

"Imperfect replication" is not what I would consider a biological term of art, so what it means depends on what is replicating, in context. Well, I'm sorry. Sometimes there's a confusion between the "official" terms of art and the phraseology some of us use to try to make these concepts more palatable.