Definition of evolution and the distinction between micro/macro

(Chris) #122

I would say even millions of years is not sufficient for this particular example of microevolution to produce even a new species; assuming you accept that as macroevolution. The reason being is that it is simply not the type of change that will accumulate into anything significant. That flock of sheep could continue for eternity to have the frequencies of alleles for white vs black wool fluctuate and they would remain sheep of the same species. It would require a different TYPE of change to produce something new.

(Chris) #123

Well technically he’s correct but most of the general public would understand “gene” but would require explanation for “allele”. I too have seen microevolution explained as a change in gene frequencies over time. Perhaps it’s best understood as an “economy” when talking to the general public.

BTW, while I understand the difference between gene and allele I am not a professional geneticist or biologist and I’m sure there are many things that would have to be explained to me in simplistic terms.

(Chris) #124

I see from your penultimate paragraph you read the links in Durstan’s article about how statistical significance and functional information could be assessed and I agree with your comment that this “has a ways to go”. I think this is only an early step and there is a lot to learn about what information is and how it is measured. Information theory really only began in 1948 with Claude Shannon and while that is a lifetime for most of us here it is really quite recently in terms of development of a scientific discipline. Even Shannon acknowledged that his measure of information had nothing to do with meaning, whereas functional information is very much to do with information.

Now regarding your stone arches. (I’ve only seen pictures but they look magnificent.) We need to distinguish between information about and information in something. It would take an enormous amount of information about the object to completely describe one of those arches. But is there “information in” the object? We might be able to derive some information about the geological processes that deposited the rock but that is not directly encoded in the rock. We might be able to derive some information about the erosion process that shaped it but again that is not encoded in the shape. On the other hand genetic information is clearly encoded IN the sequences of bases in functional DNA. There is a code and we can read a codon and match it to an amino acid.

Similarly we could produce information about a stretch of DNA which might truly be junk and not encode any biological information. Even describing it by the CAGT base sequence is still only a shorthand notation and like the stone arch we could could produce an enormous amount of information describing the position and type of every atom in the sequence but still not increase the information it encodes.

Now in terms of Durston’s definitions I could see that a number of statistically insignificant changes could add up to produce a statistically [significant] one. However I dispute that this will inevitably happen. In my parable about the sheep I tried to show such a case. No matter how long the process is extended, even over millions of years, the minor changes in wool colour will not add up to macroevolutionary change.

(Chris) #125

But if we want to communicate with someone else the word has to mean the same thing for both parties or else communication fails.
Sometimes we use a word in one sense which is understood in another sense by the hearer, and so we are misunderstood.
Sometimes we can intentionally use a word in one sense knowing that it will be understood in another sense by the hearer, and so we intentionally mislead them.

(Chris) #126

That’s a good point. In Durston’s definition losing information (for limbs, hair, and other features) would count as microevolution. Gaining information (for limbs, hair, and other features) would count as macroevolution.

Sometimes an organism can lose functional information and gain a beneficial result. A bacteria might lose the functional information to produce a particular channel in the cell membrane but if this is used by an antibiotic this can make the bacteria resistant to that antibiotic.

However it can’t make progress by continually losing information. Eventually there is nothing left to lose and still survive. Hence I think Durston’s specifying a gain of information is appropriate.

(Phil) #127

Or would it? In the case of the sheep, the farmer only bred from one ram, narrowing the genetic diversity, and with neutral drift, one one expect change with time. You can look at breeds of sheep or pigs from 100 years ago, and the animals you see on farms now are seldom the same breeds as they once were. Of course, other selective characteristics have been bred for, but I suspect even without selective breeding,you would see changes. Those changes would be cosmetic for the most part at first, but could add up over time to form a sheep that would not and could not breed with its ancestors.
Perhaps I am misunderstanding your point, but it seems you are saying the system is static other than fleece color, when it never is.

(Chris) #128

I am concentrating on one particular allele/trait. By your comment introducing other factors it would seem that you agree that changes in fleece colour alone would not result in any macroevolutionary change, over any time period.

(Jon Garvey) #129

Generality is the point, Merv. A narrow definition with undefined exceptions is either defining the wrong entity, or needs to be broadened.

If you take a legal definition (your example), it tends to be based on “what a reasonable person would conceive as X or Y”, with a human judge deliberating on the rights and wrongs of a messy situation.

A definition in science is supposed to produce precision and clarity, supposedly matching something definite in the real, objective, world. And yet it ought to be flexible enough to negotiate changes in theory.

And so gravity has a simple definion related to the attraction of massive bodies. If it had been defined according to corpuscular theory, or Newton’s theory of pressure in a medium, or even as deformation of the space-time contiunnum, each would have excluded the others and we would not see that Newton and Einstein are explaining the same, easily comprehensible, phenomenon. As far as I know, anything we know about gravity is covered by the simple, but comprehensive, definition.


(Lynn Munter) #130

Even DNA which is currently useless has no guarantee that it was useless in the past or will continue to be useless indefinitely into the future. Novel proteins might occur after a frameshift mutation, or it might be swapped into the middle of another gene to create a new gene, or it might encode something which floats around without doing anything but maybe something else will happen in the future and interact with it. None of these are too improbable on the grand scale of how many cells are living in the world. There is not a hard and fast barrier between junk and not junk.

This is part of the problem with his definition I was trying to show, for sure. The other part is that say we have an increase in information, but it is not ‘functional.’ And then say we lose information such that the existing information becomes more functional. Where did the macroevolution occur? It is a multi-step process, and failure to recognize this makes it very frustrating to hear people proclaiming “macroevolution has never been observed!”

I am afraid any collection of sheep on a farm would eventually evolve new species in the normal course of things. Color may be one part of it, but you can’t keep all kinds of other mutations from showing up.

The question of whether they would evolve into non-sheep is a little more interesting. We could argue that a dog is still in some sense a wolf, that they and foxes will always be canids no matter how far they evolve, just like they will always be carnivora and mammals and vertebrates. You don’t lose your position no matter how much evolution occurs. In this sense birds are still dinosaurs and we are still apes and primates as much as we are tetrapods. But some words are so strongly defined in ways that exclude certain categories that it just plain sounds funny to make this sort of claim: for example, you’ll likely argue when I tell you we are all still fish.

Anyway, the sheep will speciate, given long enough under real conditions, but not into non-sheep, even if they become unrecognizable as anything we would call a sheep.

(Matthew Pevarnik) #131

That seems non-scientific to me. Let me explain though- how does one decide what counts as a gain vs. loss? Generally speaking this definition seems to be contrary to how anyone actually uses the terms – it almost appears to me as if it is a made up definition of both micro and macro evolution. But let’s try to get specific with the question: does it represent an overall gain or loss for the ancestor of a whale to gain the ability to see underwater?

(Chris) #132

The links in his article describe a method of measuring gain/loss in some cases. I’m sure this will not be the one and only method but it is a step forward.

Aren’t they all? No definition is perfect but Durston has made an attempt to define them on objective grounds. Can you do better?

I don’t know; and that’s assuming that whales evolved from a land animal rather than being created.

(Chris) #133

That’s a faith statement. There is no guarantee that they will become a new species. They might, but I would agree not into non-sheep.

Remember that YECs believe that rapid speciation occurred after the animals left the Ark. Far faster than the rate of speciation we observe today. The ovine kind probably speciated into sheep and goats, and further speciation could not be ruled out. Speciation today appears to be rare although the London Underground Mosquito is a candidate. (The Talk Origins collection of examples of speciation has not stood up to examination.)

Dogs, wolves, and coyotes can still interbreed so they are almost certainly the one kind. Foxes are probably part of that kind even though I know of no hybrids with the others.
[edit 6/8/2018: apparently the coyote and red fox can interbreed]
Similarly in the cats there is a chain of hybrids linking tabby to tiger so still, in some sense, all part of the cat kind.

(Chris) #134

I can’t, but that is not the same as saying that changes in allele frequencies result in speciation.

I would hypothesise that change in allele frequencies is a necessary but not sufficient condition for speciation.

(Matthew Pevarnik) #135

So Durston provides a link to a paper that he wrote with a few others that are involved in the ID movement. That’s fine, though I was curious who was actually citing his paper titled Measuring the functional sequence complexity of proteins, finding 8 citations. I typically look at citations of papers in fields I’m not that familiar with to see at least its impact on the scientific community. I see a few citations by Dembski, Ewert, and Abel who all obviously would be sympathetic to some impossibilities of ‘macroevolution.’ It is obvious how such writers will engage Durston’s ideas but what is more interesting to me is how the scientific community engages his ideas who doesn’t already have a vested interest in his ideas being true. However, one of the papers citing Durston’s method appears to not be sympathetic to ID, the authors come to a different conclusion altogether:

His definition is pretty much a glorified (perhaps he can correct me if he pops in here):

Sure, the language is clouded in some probability arguments that permit a certain amount of change but not too much! My question to you above was ‘how far can microevolution change a population?’

Also, perhaps what I meant to say is not that I can come up with a better definition, but rather he is making up definitions independently and somewhat contrary to how they are used in the biological literature as a whole (even though the two phrases are quite rare in general).

Let me go back to this quote of your from above for a second:

I asked because genomes and morphologies speak to particular histories of various species. We can ask what kinds of genetic changes would have to result to get from a land dwelling mammal to an aquatic one. That to me, appears to be a very large ‘macroevolutionary’ change so to speak. Here is a neat paper that asks, are there signatures in the genome that speak to such changes occurring and they find that many of them were quite small! Just like someone who has a blueprint that they make a few tweaks to each iteration:

(Matthew Pevarnik) #136

Perhaps you can elaborate on this a bit. As in it appears that you are well extremely close to well agreeing with evolution. This is where one eventually aims to identify the original ‘kinds’ but the original ‘kinds’ are more similar that everything they supposedly branched in to (and all of the ‘kinds’ should be their own ‘kind’ and so forth until one accepts common descent of all species).

We have two options here. Either the rate was unbelievably high in the past and then all of a sudden slowed down to virtually stop with no mechanisms or methods proposed (as you seem to be advocating) or the rate is actually indeed quite slow–as in speciation is ‘rare’ on the timescales we can observe quite naturally. The most rapid speciation documented for marine vertebrates (as of 2017 that I’m aware of according to the abstract) took 8,500 years:

(Chris) #137

Perhaps another way of looking at macroevolution is to say this the type of evolution that transformed microbes to man. This would require the development of many new features and functions. Then we could give a definition as;
Macroevolution is evolution that produces new features or functions.

This is close to Futuyma’s definition; “A vague term for the evolution of great phenotypic changes, usually great enough to [put] the lineage into a distinct genus or higher taxon.”


The point of using allele instead of gene is that we are following a mutation through time in a population. It only makes sense to compare the mutation to the unmutated section of DNA (i.e. orthologous DNA). For example, if there is an MHC allele that confers resistance to small pox it would make sense to follow that allele since everyone will have the MHC gene in question.

That definition works for most situations, but it may not work as well in cases of horizontal genetic transfer.


I don’t see how that would work. If we compared the chimp and human genomes, would we see expect to see a difference in functional information? Could macroevolution simply be a shift in function that would still have the same amount of overall functional information?


What about the evolution of primates (including humans) from a shared common ancestor? Would that qualify as macroevolution within this regime?

The line between genera is an arbitrary human construct, so that is a very subjective definition (as I am sure Futuyma would agree). There is no objective criteria for putting species in the same genera.


Durston’s method only looks at protein sequences, not at morphological change. I think it is wrong to assume that you need new proteins in order for a feature to change. Is the evolution of a fish fin into a tetrapod limb an increase in functional information?

On top of that, Durston’s method does not take promoter and RNA sequences into account. Changes in gene expression are often as important as changes in protein sequence, and Durston’s method will not necessarily pick up on these changes. It is also interesting that Durston’s method seems to be saying that the only functional information in the genome is found in translated proteins which would put the percentage of functional DNA in the human genome at about 1-2%.