Speciation in Evolution

[SezMe]

I'm not a biologist but I thought this article was interesting. I'd be interested in comments from those more knowledgeable than I.

 

[marplots]

It doesn't seem controversial to me, especially when you add extinction into the mix as a confounding factor. All it depends on is "soft" niches - so many available ways to live that survival doesn't depend on a particular ability.

My reading is that it's just another way of looking at what everyone already agrees on. But I'm an amateur here and I await my betters and their comments.

 

[Gawdzilla Sama]

It doesn't seem controversial to me, especially when you add extinction into the mix as a confounding factor. All it depends on is "soft" niches - so many available ways to live that survival doesn't depend on a particular ability.

My favorite example is the zebra. One zebra may have a better pattern than others, but not be the fastest zebra in the herd.

(And, yes, I've heard of the various opinions about what function the stripes actually serve, it's just an example, not a definition.)

 

[jonesdave116]

I think the controversial part of the hypothesis is the 2 million years per speciation event, regardless of generation time. They're effectively saying that a mosquito will throw up a new species, given geographical isolation, no faster than an elephant. Not sure about this.

 

[ryu238]

...it's ironic, we've told creationists that evolution is partly random...

 

[marplots]

I think the controversial part of the hypothesis is the 2 million years per speciation event, regardless of generation time. They're effectively saying that a mosquito will throw up a new species, given geographical isolation, no faster than an elephant. Not sure about this.

That's a nuance I didn't pick up on, but I could see why generation-time wouldn't necessarily make a difference. If mutation rates (here I mean successful, species-defining types of mutations) are relatively slow, then the generation-time would impact the spread of the mutation, but the overall rate could still be the driver - especially if it takes 2 million years for a suite of chance mutations to have a significant impact.

In the extreme (for illustration), say there's only one worthwhile mutation every million years. Then, it doesn't matter if you are a mosquito or an elephant, the event is going to impact your species the same, regardless of how fast you breed. So long as your generation-time isn't on the order of hundred thousand years between births.

This seems like a math question to me. Still, I would like to hear more in the way of explanation/explication.

 

[Beerina]

So basically the vast majority of DNA changes are little quirks that don't affect the organism, and that a buildup of these is what causes the speciation (DNA is no longer compatible) milestone to occur.


I always thought sexual reproduction, although somewhat controlled (compared to random neutrons & friends) did the vast majority of changes, and that these could build up at the DNA level as well (e.g. you aren't just safely touching some quasi-variables leading to, say, leg length). Sexual reproduction is magnitudes faster in scouring the evolutionary fitness gradient descent space, which is why it itself evolved, leaving "only" chance nuclear or chemical copy error mutation far behind in the dust.

 

[Swordfishtrombone]

As is so common in these sorts of articles, if you only read the title and short blurps, like here "Researchers build the world’s largest evolutionary tree and conclude that species arise because of chance mutations — not natural selection.", you get the impression that what is being proposed is totally radical, and about to turn everything you know on it's head.

Of course, when you read further, and actually find out what it's about, you find that what is being proposed is far more modest, but still interesting.

So essentially, one point they are making as if it was new is that while geographical factors and differing changes in the separated populations driven by natural selection can start the process, that these changes don't alone lead to speciation, but random mutations acquired over time in the separated populations gradually make them genetically incompatible, and incapable of interbreeding.

I thought this we already knew? At least this is how I've understood speciation to work.

So what is new is the assertion that for this sort of a speciation event to reach it's conclusion, it takes about two million years, on average, regardless of what species we are talking about.

And that IS really interesting, if true.

 

[Lowpro]

I think the controversial part of the hypothesis is the 2 million years per speciation event, regardless of generation time. They're effectively saying that a mosquito will throw up a new species, given geographical isolation, no faster than an elephant. Not sure about this.

Not quite how to interpret that. The base rate of mutation can best be predicted at the same rate between all species. That's just the chemistry of it. If you're at a HWE (sans noxious mutations) then yes, elephants and mosquitoes will alter their genes through mutations at the same rate.

But HWE doesn't really occur in reality. None of this really is controversial. Sewell Wright pretty much figured all this out so I'm surprised that anyone could find much controversial here. All Hedges did was expand the Sewell Wright model. Speciation is a genetic quality and natural selection influences the genes that get to play in the sandbox. Even without any obvious natural selection (ie at HWE) you'll still have mutations enough to have a different species given time.

So wherever you are Dinwar and Kotatsu, what's your thoughts on this article? Clickbait?

 

[barehl]

I'm not a biologist but I thought this article was interesting. I'd be interested in comments from those more knowledgeable than I.

The 2 million year clock seems about right for gene drift. That would be about 100,000 generations for a species with a 20 year generation. That would work for humans but not for rats or horses.

 

[Trakar]

So how does this research accommodate ring species?

 

[Craig B]

That's a nuance I didn't pick up on, but I could see why generation-time wouldn't necessarily make a difference. If mutation rates (here I mean successful, species-defining types of mutations) are relatively slow, then the generation-time would impact the spread of the mutation, but the overall rate could still be the driver - especially if it takes 2 million years for a suite of chance mutations to have a significant impact.

Is it radical mutation that is the main driving force of evolution? I thought separation of populations within the same species played a leading role. Some S American tortoises found their way to the Galápagos Islands and new species gradually evolved there.

The evolution did not require the S American tortoises to await a radical gigantism mutation for the species now on the Galápagos to be formed.

 

[MG1962]

Is it radical mutation that is the main driving force of evolution? I thought separation of populations within the same species played a leading role. Some S American tortoises found their way to the Galápagos Islands and new species gradually evolved there.

The evolution did not require the S American tortoises to await a radical gigantism mutation for the species now on the Galápagos to be formed.

I was wondering the same thing about the same example. Every isolated island a species of tortoise has made it to has seen some variation of gigantism occur

 

[The Shrike]

"Consider Hawaii’s honeycreepers. The speciation clock started once the birds migrated to a new island and began to accumulate random mutations. The vast majority of these mutations were neutral, having no effect on the birds’ appearance or behavior. "

There's a bit of a rub here as well. A mutation is neutral if it has no particular bearing on fitness, i.e., lifetime survivorship and reproduction. There are all sorts of mutations that could do something to affect phenotype without affecting fitness. If the thing done to the phenotype leads to reproductive isolation (e.g., a change in a bird's song to which females from the "parent" population don't respond), then yes, speciation can be driven by genetic drift in the absence of strong selective pressure.

 

[marplots]

Is it radical mutation that is the main driving force of evolution? I thought separation of populations within the same species played a leading role. Some S American tortoises found their way to the Galápagos Islands and new species gradually evolved there.

The evolution did not require the S American tortoises to await a radical gigantism mutation for the species now on the Galápagos to be formed.

I was wondering the same thing about the same example. Every isolated island a species of tortoise has made it to has seen some variation of gigantism occur

It's the same thing, isn't it? If we never separate a subgroup, then any neutral or positive mutations permeate the entire population. You still get different species, but they are only different through time, not contemporaries.

However, all separated subpopulations moving in the same direction (the island-separated tortoise giantism) would argue against this, unless it was both a common mutation and something with a huge survival advantage.

I wonder what they have to say about animals that don't seem to evolve much over really huge blocks of time - Crocodiles over 200M years? The current explanations I've heard:
1) They did change, but there weren't any dramatic phenotype changes picked up in the fossil record - leaving immune systems and other targets instead. Don't know how I feel about this one. Hard to prove, I guess.
2) They have a mechanism to deal with mutations that is more powerful than the norm. Not sure about this either.
3) Their niche is so specific, it tends to extinguish change. Again, seems like something you'd have to prove, but this argues against the "mutation-driven" scenario, which would rely on weak niche influence.

Speculation is fun, but I'd like to hear from Dinwar.

 

[Dinwar]

People are surprised by this? This was stuff I learned as an undergrad, right along with how to prep fossils and how to identify the various phyla.

As for the rates, I'm not sure I believe it. There's some pretty interesting controversy over how to apply genetic data to the fossil record, and many studies that attempt to tie genetics to fossils have a poor understanding of calibrations in the fossil record (that's assuming they even look at fossil calibrations, which is often not the case). I take all studies where biologists say "Genes do this through time" with a HEAVY grain of salt, for the same reason anyone should take a paleontologist saying "Modern organisms do X" with a pretty heavy one.

Is it radical mutation that is the main driving force of evolution? I thought separation of populations within the same species played a leading role.

You can separate populations for a billion years, but if the genes don't change they'll still be the same species. Mutations are necessary to drive differentiation between the species. They don't have to be radical individually, but en mass small changes can have profound effects.

 

[MG1962]

It's the same thing, isn't it? If we never separate a subgroup, then any neutral or positive mutations permeate the entire population. You still get different species, but they are only different through time, not contemporaries.

However the common wisdom seems to be, the bigger the base population the slower mutations have an effect over the entire population. Islands present a fascinating laboratory for genetic variation because it is highly likely the entire genetic make up of a species is not represented. So in theory mutation and the rise of new species should be a lot faster in such and environment. But if I am reading this article correctly. That is not the case at all.

 

[marplots]

However the common wisdom seems to be, the bigger the base population the slower mutations have an effect over the entire population. Islands present a fascinating laboratory for genetic variation because it is highly likely the entire genetic make up of a species is not represented. So in theory mutation and the rise of new species should be a lot faster in such and environment. But if I am reading this article correctly. That is not the case at all.

I see what you mean about bigger base populations, but I assume that's swamped by the two-million-year step. Provided, of course, that 2MY is a reasonable number in the first place.

 

[Dinwar]

The real issue here isn't "Is this true or not?" It almost certainly is true that mutations play a huge role in determining the evolutionary history of a species--in some cases. In others, the niche will play a more significant role. It's a question of which is dominant in the taxa in question.

Take marine predators, for example. Icthiosaurs, dolphins, sharks, and submarines all have the same body shape, despite being very, very different taxa (and one group not being biological at all). That's because of very serious physical constraints--fluid dynamics demand that if you want to move fast through the water, you will be shaped in a specific manner. In contrast, browsing is pretty easy, and as long as you have a way to get the plant matter into your digestive system you can have any morphology you wish. So it's pretty clear that there are multiple ways for evolution to work, not just the One True Method.

 

[SezMe]

People are surprised by this? This was stuff I learned as an undergrad, right along with how to prep fossils and how to identify the various phyla.

Lighten up, Dinwar. I didn't say "surprised", I said I found it interesting.

I went to a high school where "evolution" didn't exist. And my entire academic career was in math and the physical sciences; I had no further biology. I do try to stay informed of the life sciences but it's not easy to keep up. Posing questions here is an excellent way to get an exposure from a variety of sources.

I'll make you a deal. I won't poo-poo your lack of knowledge of numerical approximation if you don't poo-poo my lack of expertise in the life sciences. OK?

 

[Dinwar]

Wasn't talking about folks here, necessarily. I am just always rather surprised to find just how deeply erroneous concepts are entrenched.

I remember talking to.a paleontologist about Punctuated equilibrium once. He said that it works for some taxa, but not others--for example, it does not work for forams. He showed me sufficient data to convince me had I been in doubt. My response was "Okay, yeah, that's obvious." Then he proceeded to demonstrate that other paleontologists were unwilling to accept this. Variability in biology is nearly the only constant--phylogeny impacts ALL of biology--yet these experts were arguing that there was One True Tempo of evolution. It struck me as absurd to the point of dishonesty on their part; we are supposed to be rational and base our opinions on data, yet these folks were willing to ignore it because of a sexy new theory that they picked up in school.

I always try to be aware of such situations, and to point out the absurdity of them. They are actually quite dangerous to science as such; history shows that such concepts can take generations to become dislodged if we are not careful.

 

[Dinwar]

Mto put it more briefly: my surprise isn't with folks here with expertise in other areas. It is with those who should know better, such as those in the article. THEY have the training to know better.

 

[The Shrike]

It is an interesting analysis, but I don't think in the way the reporting suggests.

First, we need some kind of definition of a species that makes some sense. None of them do except with respect to our human desire to attach discrete names to things. The whole of life is an evolutionary continuum. We humans can notice a "boundary" at nodes on a branch where gene flow is restricted over time. If the loss of gene flow is dramatic or prolonged enough, then differences that accumulate over time can be sufficient that prevent any renewed gene flow. That can occur through little changes over a long time (genetic drift) or big changes over a short time (natural and/or sexual selection).

We have abundant examples of very rapid diversification in a lineage - adaptive radiations - and they are very well studied on archipelagos. The combination of isolation from parent stock and intense natural selection in the novel environment is a recipe for speciation, and it don't take no 2 million years. (By my count, there are 5 extant endemic bird species on Hawaii, which is estimated to be 500,000 years old).

I suspect what Hedges and his team have encountered is speciation (largely judged indirectly through gross morphological changes apparent in the fossil record) occurring without isolation and/or marked changes in natural selection. In other words, a raccoon today will likely be quite different from its distant descendants, just through the accumulation of random mutations over time. How long does it take for descendants and ancestors to diverge in this way? Sounds like about 2 million years.

If that analysis holds up, it is in fact really cool. To my knowledge, no one has ever suggested that X amount of genetic change = Y likelihood of speciation. Because the great majority of living things do not occur in extreme environments conducive to rapid speciation, a "whole biosphere" analysis would probably result in Hedges' genetic drift over time thing swamping the influence of things like adaptive radiations in the grand scheme of the development of life on Earth.

 

[Dinwar]

A fairly minor change can result in speciation, if it is the rigt change and the right species concept. A shift in mating times, for example, can preclude mating between two populations--and result in new species by the biological species concept.

This still doesn't address asexually reproducing species, or really anything outside of Animalia.

We have case studies of speciation absent isolation: they are called chronospecies. They throw a nice monkey wrench into the naive assumptions about the term "species". Not sure how frequently bison speciated; once I find the relevant texts I may have to look that up.

Also, I take umbrage at the dismissal of morphospecoes as a valid definition. Morphology takes into account far more genetic material than most genetic studies,cas morphology is controlled by numerous genes, not just those selected,for analysis by a researcher. Studies have shown that morphological and genetic analysis yield the same results (witchin given tolerances, obviously), meaning that you can't say one is more precise than the other. And genetic species concepts are not inherently superior to the morphospecoes concept, they just are applicable in different situations. The morphospecoes concept is the workhorse of biology.

 

[The Shrike]

^Dude, you were born taking umbrage. I'm dismissing all species concepts, not ranking them relative to one another. I'm well aware that the morphospecoes [SIC] concept is essential to our understanding of life on Earth.

The chronospecies is what I think might be driving Hedges' analysis, but I'll need to read the original before I climb too far out on this branch.

 

[ryu238]

So basically the vast majority of DNA changes are little quirks that don't affect the organism, and that a buildup of these is what causes the speciation (DNA is no longer compatible) milestone to occur.


I always thought sexual reproduction, although somewhat controlled (compared to random neutrons & friends) did the vast majority of changes, and that these could build up at the DNA level as well (e.g. you aren't just safely touching some quasi-variables leading to, say, leg length). Sexual reproduction is magnitudes faster in scouring the evolutionary fitness gradient descent space, which is why it itself evolved, leaving "only" chance nuclear or chemical copy error mutation far behind in the dust.

Interesting, what do you think of this then?
http://creation.com/refuting-evolution-2-chapter-11-argument-evolution-of-sex

 

[Dinwar]

^Dude, you were born taking umbrage.

Not really. You just have a bad habit of saying things that are not true,then declaring that you didn't say them when called out on it. Or denying the whole conversation. Whatever; not really interested in another futile conversation laced with insults, and I will not peruse this further.

For the record, it was your parenthetical in the third paragraph that is the issue here. The way it is written, you appear to be saying that morphology is only an indirect indicator of speciation. This can only mean that you subscribe to a strictly genetic or biological species concept, or a combination of the two.

 

[Nonpareil]

Interesting, what do you think of this then?
http://creation.com/refuting-evolution-2-chapter-11-argument-evolution-of-sex

Full of complete idiocy and written by someone with no understanding of the subject. I got to this point:

Indeed, the program acknowledges that sex has many disadvantages, e.g., only 50 percent of the genes are passed on to an offspring. This means that there is a 50 percent chance of losing a beneficial mutation.

...and stopped reading.

 

[marplots]

It is an interesting analysis, but I don't think in the way the reporting suggests.

First, we need some kind of definition of a species that makes some sense. None of them do except with respect to our human desire to attach discrete names to things. The whole of life is an evolutionary continuum. We humans can notice a "boundary" at nodes on a branch where gene flow is restricted over time. If the loss of gene flow is dramatic or prolonged enough, then differences that accumulate over time can be sufficient that prevent any renewed gene flow. That can occur through little changes over a long time (genetic drift) or big changes over a short time (natural and/or sexual selection).

We have abundant examples of very rapid diversification in a lineage - adaptive radiations - and they are very well studied on archipelagos. The combination of isolation from parent stock and intense natural selection in the novel environment is a recipe for speciation, and it don't take no 2 million years. (By my count, there are 5 extant endemic bird species on Hawaii, which is estimated to be 500,000 years old).

I suspect what Hedges and his team have encountered is speciation (largely judged indirectly through gross morphological changes apparent in the fossil record) occurring without isolation and/or marked changes in natural selection. In other words, a raccoon today will likely be quite different from its distant descendants, just through the accumulation of random mutations over time. How long does it take for descendants and ancestors to diverge in this way? Sounds like about 2 million years.

If that analysis holds up, it is in fact really cool. To my knowledge, no one has ever suggested that X amount of genetic change = Y likelihood of speciation. Because the great majority of living things do not occur in extreme environments conducive to rapid speciation, a "whole biosphere" analysis would probably result in Hedges' genetic drift over time thing swamping the influence of things like adaptive radiations in the grand scheme of the development of life on Earth.

I have a suggested answer to the question of, "If speciation-style mutations are constant (the two-million-years number), then why do we see such rapid speciation in extreme environments?"

My suggestion is that the difference between neutral mutation and positive, speciation-causing mutation, isn't necessarily the mutation itself, but the environment as well. In a much-different, high stress environment, a neutral might actually be a positive.

But this is just niche-matching and survival of the fittest found in "regular" evolution anyhow.

A neat question, which has likely been addressed by those in the field, but which I don't know the answer to, is whether there is a fixed background to variability/mutation rate, or whether the mutation rate itself is variable and subject to feedback. And here I mean genetic error/mutation rate, not the animal which emerges because more of the fixed-rate mutations are beneficial.

Dinwar? Can you shed some light on this question of mine?

 

[Dinwar]

My suggestion is that the difference between neutral mutation and positive, speciation-causing mutation, isn't necessarily the mutation itself, but the environment as well. In a much-different, high stress environment, a neutral might actually be a positive.

Part of the problem, as illustrated by chronospecies, is that the term "speciation" isn't precise. It's not a thing, not a process, not really a suite of processes--it's an end result, which can be brought about by any number of processes.

That said, more extreme environments will have more extreme fitness spaces. That's what this all boils down to; in some cases, the topography of fitness spaces (ie, the selection pressure) is such that randomness in genetics dominates, while in others the topography is such that selection dominates. Random mutations, genetic drift, etc., are universals in biology--they ALWAYS happen--so the question is, how extreme does selection have to be to dominate?

A neat question, which has likely been addressed by those in the field, but which I don't know the answer to, is whether there is a fixed background to variability/mutation rate, or whether the mutation rate itself is variable and subject to feedback.

Tricky to determine. The problem is, a LOT of selection is in-utero. A lot of mutations are such that the zygote doesn't survive long enough to get noticed, much less be born. A mutation to the HOX Box can easily prove fatal, as can mutations to key proteins. So we're left with the rather unenviable option of looking at mutation rates in surviving organisms. That said, we can look at unexpressed genes, or genes where mutations CAN survive, to get a sense of this.

I know the assumption is that mutations rates in molecular clocks are stable over geological periods--ie, they have a constant rate. I also know that some bacteria appear to be able to increase the mutation rate in some situations, so it's almost certainly not as simple as "mutation rates are constant". And the real problem, to my mind, is that we can at best only have observed mutation rates for fifty years or so, which does not allow for highly accurate extrapolation into geological timescales.

So the short answer is we have a few good guesses, but we're working on it.

 

[marplots]

So the short answer is we have a few good guesses, but we're working on it.

This actually pleases me greatly.

And thank you for the informative response.

 

[Darwin123]

Is it radical mutation that is the main driving force of evolution?

Radical mutation is not a significant driving force for evolution. The mutations that contribute to evolution have phenotypic effects that are barely perceptible. A population evolves due to the accumulation of barely perceptible changes, all of which originate as de nova mutations that are barely perceptible.

The mutations are random with respect to the fitness of the organism. Therefore, mutation without selection merely increases the variation of a population without leading to a new species. A 'driver' by definition has to set a direction. Therefore, the mutations themselves do not drive evolution. Natural selection drives evolution because it determines the direction.

Natural selection holds the steering wheel, so it is the driver. However, random mutation may have its foot on the accelerator. There is a controversy whether the rate of mutation correlates with the rate of evolution.

The rate of de nova mutations has to be fairly high for evolution to occur. However, the rate of evolution has to also depend on natural selection. Both on the rate the environment changes and on the degree of inhomogeneity in the environment. Since the environment changes at differing rates, most scientists expect that evolution occurs at differing rates.

The linked article is not claiming that evolution occurs through 'radical' mutation. The author is assuming that evolution occurs through an accumulation of very small mutations. However, he claims that the average time for a species to form is fixed. If this is true, then it would show that the rate of evolution is largely fixed by the rate of de nova mutation.


This hypothesis is almost fantastic. However, it is rational enough to hold my attention. Most studies that I have read suggest the opposite. I will reserve judgement until more studies are done.

Punctuated equilibrium theory says that the rate of evolution is almost entirely determined by changes in environment. In fact, evolution occurs mostly after the environment changes catastrophically in a very short time. Evolution when the environment is stable is very slow.

Evolution may be slow when the environment is stable. However, how slow is it? A lot of evolution can occur in the stable periods, even if it speeds up after a disaster.

Maybe only speciation occurs immediately after a mass extinction. That is my impression. Punctuated equilibrium applies for taxonomic orders ranging from species to family. However, I don't think higher taxonomic orders are created immediately after a mass extinction. According to what I read, orders and classes of large organisms take millions of years. So I might believe it if someone said that the rate of de nova mutation determines the rate that orders evolve. I can't believe that the rate of mutation is really significant on the species level.

thought separation of populations within the same species played a leading role. Some S American tortoises found their way to the Galápagos Islands and new species gradually evolved there.

The evolution did not require the S American tortoises to await a radical gigantism mutation for the species now on the Galápagos to be formed.

I don't think the author, or any other scientist, believes that gigantism in tortoises occurred through one radical de nova mutation. The tortoises on separate islands probably grew through a series of mutations where each contributed a small increment in size.

The article seems to be suggesting that gigantism is common because the rate of random mutation is so big. If that is so, gigantism on each island occurred through different de nova mutations. This would mean that different genes mutated on different islands even though the accumulated effect was the same.

Scientists believed gigantic tortoise species in the Galapagos evolved independently. This is called convergent evolution. If so, each tortoise species would have a different genetic reason to be gigantic.

The de nova mutations leading to a gigantic tortoise would be different on each island. No one is claiming that one radical mutation caused gigantism on each island. What the article suggests is that maybe the same number of de nova mutations were necessary to create gigantism regardless of the exact sequence.


If the same genes lead to gigantism, then I would say that the giant tortoise evolved on one island and spread out. It differentiated after it spread out.

Just for clarification: de nova mutation does not mean radical. A de nova mutation is the first spontaneous change in a gene. The de nova mutation is never inherited, it is spontaneous. If the modified gene is inherited by offspring, the offspring does not have a 'true' mutation. However, people have kept misusing the word 'mutation' so many times that I feel that a qualifying phrase is necessary.

Recommendation to everyone.

Please use the phrase 'de nova mutation' when you are referring to the spontaneously modified gene before offspring inherited it. Also try to avoid the word 'mutation' when referring to a modified gene that was inherited from parents. It doesn't matter whether the effect of this gene is large or small. Only the spontaneous (i.e., first) change is a mutation.

The phrase 'rate of mutation' will be interpreted by me to include ONLY de nova mutations. I think it would help others if this convention is adhered to.

 

[Dinwar]

The mutations that contribute to evolution have phenotypic effects that are barely perceptible.

Depends on the mutation. Some are barely perceptible. Some are glaringly so. That was a major finding of de Vrise' primrose experiments, and what allowed him to propose the existence of mutations.

Punctuated equilibrium theory says that the rate of evolution is almost entirely determined by changes in environment.

Sorry, no. It states that the rate of evolution is determined by fitness space, which can be related to changes in the environment. In fact, Gould and other PE advocates argued that most organisms will follow the environment they prefer (if possible, over generations), mitigating the effects of the environment on rates of evolution. That's part of stasis.

In PE, the majority of evolution occurs at speciation events. These can be acused by a reduction in selection pressure (ie, the aftermath of some catastrophy), or due to random mutations/genetic drift bringing a population across a fitness valley, or any of a number of other things. It's about the timing of evolution, not the cause.

In fact, evolution occurs mostly after the environment changes catastrophically in a very short time.

A common misconception. In fact, most speciation occurs without mass extinctions; mass extinctions account for only a small amount of new speciation. Local and regional effects are less well-understood. There have been six, maybe seven mass extinctions (depends on how you count). Those cannot account for all diversity, or even a significant chunk of it.

 

[The Shrike]

Not really.

So you're taking umbrage to the charge that you're quick to take umbrage? Got it.

You just have a bad habit of saying things that are not true,then declaring that you didn't say them when called out on it. Or denying the whole conversation. Whatever; not really interested in another futile conversation laced with insults, and I will not peruse this further.

I'd ask what the Sam Hill you're talking about, but the most parsimonious explanation seems to be that you've confused me with someone else (with whom you've also taken umbrage).

. . . you appear to be saying that morphology is only an indirect indicator of speciation. This can only mean that you subscribe to a strictly genetic or biological species concept, or a combination of the two.

Nope, I see them as all equally valid concepts, which renders them equally invalid as well.

Using a morphospecies approach, the variability in size and bill shape would be enough to identify multiple species within the currently recognized Song Sparrow. With the benefit of more recent biological material (e.g. plumage) than tiny fossilized skeletons, researchers have described at least 25 different subspecies. In this case, one species using biological or genetic concepts becomes multiple species using a morphospecies concept.

In contrast, there are about 33 species of Setophaga warblers that are all quite distinct by plumage, behavior, song, etc. but if all we had to classify them were fossilized skeletons, I doubt all 33 would be recognized.

 

[Dinwar]

I'd ask what the Sam Hill you're talking about, but the most parsimonious explanation seems to be that you've confused me with someone else (with whom you've also taken umbrage).

If there is a second person named "The Shrike", you should report them. They are rapidly eroding your credibility. If not, again, I've drawn my conclusion, and this only supports it.

Using a morphospecies approach, the variability in size and bill shape would be enough to identify multiple species within the currently recognized Song Sparrow.

Common misconception. The reality is that each taxa has specific traits that are diagnostic at each taxonomic level. Certain traits seem to vary most within species, others between species, others between genera, others between families, and so on. It takes years to understand which traits tell you what about each organism. This makes sense, however; higher-order taxa evolved earlier, and therefore SHOULD be identifiable via different markers than lower-order taxa, and different organisms have different responses to selection pressures and statistical genetic issues. My point is, it's not intuitive which characters are important. It requires a fairly intimate knowledge of the taxa in question to understand this.

All that said: I don't know much about the characters in Avis. I would be astonished, however, to find that variability in bill shape is so important. I'm not entirely convinced that dental variability is as critical as often portrayed in Mammalia (though I am in the minority here).

All THAT said, yeah, it's trivially obvious that when going from one species concept to another there will be times when multiple species are combined into one, or one is divided into multiple. For that matter, any time you work outside Animalia the concept of species becomes problimatic, and within Animalia it's iffy (how do you define the species of a tapeworm?). The issue is, "species" do not, as such, exist in the wild. The concept of a species is a human-made construct that assists in understanding biology. What exists in biology are populations. The idea of uniting various populations that do not and cannot interbreed into one over-arching heading because they could, after being ripped from their habitat and brought together with something completely alien to them, successfuly produce offspring is really pretty strange when you think about it. As such, I see no reason to assume one species concept is inherently superior to any other--and that means that we can move between them as needed without any hesitation, so long as we acknowledge the limitations of each concept.

I will say that limiting yourself to osteological evidence is on you, not paleontology. We work with squishy bits quite frequently. Not as frequently as we'd like, but far, FAR more than most realize. We've actually disected organisms that have been dead 540 million years. Those papers are at the same time the most facinating and the most god-awful boring things to read that I have yet encountered.

And if you REALLY want to attack the morphospecies concept, and paleontology, look at worms. If every worm went extinct today there'd be no fossil record of it, except for the near-total sterilization of the planet.

 

[Darwin123]

Depends on the mutation. Some are barely perceptible. Some are glaringly so. That was a major finding of de Vrise' primrose experiments, and what allowed him to propose the existence of mutations.

I specified 'mutations that contribute to evolution'. Mutations that are 'glaringly perceptible' are not going to contribute to evolution. I know that radical mutations occur. Saltations very often occur. However, the mutations that accumulate are usually those which are barely perceptible.

De nova mutations that are very large tend to overshot their range of usefulness in fitness space. However, these are the mutations that are least likely to be missed by observers. So the existence of mutations was first demonstrated by radical mutations, the very mutations that can't accumulate.

De Vrise may have demonstrated the existence of mutations with his primrose experiments. However, he did not demonstrate that the contributed to evolution. He observed radical mutations, which could also be called saltations. However, these saltations would have a difficult time propagating in a natural environment. The habitats where these saltations would be at an advantage are very rare.

I read an article which I don't keep handy. However, the article showed an experimental study which demonstrated that there are on average 30 de nova mutations per human birth. Obviously, most of these mutations are barely perceptible. The study was based on sequencing of genomes. The phenotypic effect of these de nova mutations was too small to detect by direct observation. There could have been phenotypic effects, but their effect on fitness would have been small (but nonzero).

Evolution would occur over many generations as natural selection acts on the genes that were modified by these de nova mutations. The fact that the effects were small greatly increases the probability that any one mutation could be an advantage in some accessible environment.

Human saltations occur much less frequently than 30 per human birth. Saltations, however, have a very high chance of being perceived by other human beings. Their chance of propagating over many generations is extremely small.

The article wasn't referring to saltations, anyway.

 

[Dinwar]

Okay, so how do perceptible mutations NOT accumulate? Remember, I am referring to a specific set of experiments here, so "frogs with nine legs don't reproduce" is not a valid answer.

Mutations will be eliminated if either they are selected against, or they are washed out via statistical phenomena. It doesn't matter what they do; those are the mechanisms.

Furthermore, you are making some VERY specific statements about what traits can evolve, without any evidence. Traits with a small number of contributing allels can, with very few mutations, produce perceptible, sometimes even profound changes. Take, for example, Nylonase. Sure, it was an extreme selection environment, but that is rather the point, after all. Not all traits are gradational, and not all graduation in characters is genetically controlled.

You are preaching Unkformitarianiam sensu stricto--a paradigm that died in the 1990s. While we can say that evolution is USUALLY slow and gradual, it is deeply erroneous to say it MIST BE so. Each trait must be evaluated independently in regards to the tempo of evolution. If not each trait, each taxa, certainly.

Boiled down, you are making a very large number of hypotheses without sufficient data to support them, and in defiance of known genetic mechanisms and experimental results. Again, I fully acknowledge that slow, gradual change does happen. I am even willing to grant, ad arguendum, that it is the most common mode for evolution. But we cannot say it is the only, or the necessary mode of evolution--because experiments show otherwise, and the mechanism demands otherwise.

 

[Dinwar]

And yes, I am ignoring the obviously inflammatory equivocation.

 

[The Shrike]

If there is a second person named "The Shrike", you should report them.

If I saw something report-worthy I would. You're the one claiming habitual lying and pointless, insult-laden conversations courtesy of this brigand. (You're also the one who sent me an angry PM recently that was clearly directed at someone else, so if there's a pattern here maybe it's you confusing the posters with whom you're interacting.)

Anyway . . .

Common misconception. The reality is that each taxa has specific traits that are diagnostic at each taxonomic level.

Of course, but I think you take this too far. Each taxa diagnostic for each taxonomic level? I think you've indicated in the past that you work with Forams, and I agree that would be the case with them - perhaps tautologically so if test morphology is the primary criterion for classification. Here's an example of the variability I'm talking about within and between two species of congeneric birds. In this case, the comparison is wing length in mm from the wrist to the tip of the longest primary:

Golden-cheeked Warbler: f 58–65, m 61–69
Black-throated Green Warbler: f 52–64, m 57–68

Golden-cheeked is a bit bigger on average, but a primary feather from a bird with wing length 58–68mm would not be assignable to species, even if correctly assigned to family and genus. The ability to discriminate them from fragmentary fossil evidence, e.g., a humerus, would be even less robust.

In life, the breeding distributions of these species do not overlap, they are morphologically distinct, and they're not prone to hybridization; they are, of course, genetically distinct as well.

So here is an example - and I'm sure you can think of many more of these than I can - in which a morphospecies concept is limiting, just as there are limitations in applying genetic and biological species concepts.

I would be astonished, however, to find that variability in bill shape is so important. I'm not entirely convinced that dental variability is as critical as often portrayed in Mammalia.

I'm with you on the Mammalia. Re: bird bills - I might agree, but can you clarify what you mean by important in your context? Important for what?

The issue is, "species" do not, as such, exist in the wild. The concept of a species is a human-made construct that assists in understanding biology. What exists in biology are populations. . . . As such, I see no reason to assume one species concept is inherently superior to any other--

Hey, this all sounds familiar. Maybe you don't take umbrage with my position after all . . .

First, we need some kind of definition of a species that makes some sense. None of them do except with respect to our human desire to attach discrete names to things. The whole of life is an evolutionary continuum.

. . . and that means that we can move between them as needed without any hesitation, so long as we acknowledge the limitations of each concept.

Wait, is that what that means? I'm not sure. This is where I need to read the paper and get a better idea of what they did. My gut tells me that they should have only included in their analysis examples using ONE strictly applied species concept, rather than jumble them all together which is what it superficially appears they did. You don't have any problem with them running the analysis on examples that applied different species concepts?

I will say that limiting yourself to osteological evidence is on you, not paleontology. We work with squishy bits quite frequently.

Very cool, but the thing that's on me is only in the context of this paper and the methods they used to obtain what might be a really provocative result. Again, I'd be a lot more comfortable about interpreting the analysis if they only used osteological evidence and only applied a morphospecies concept. Your experience tells you that I'm being too rigid and that my concerns over source material and different species concepts are red herrings in this context. Correct?

Whether you believe me or not, I enjoy reading your perspectives on evolution Dinwar, and I very much appreciate you taking the time to share them.

 

[Dinwar]

I think you've indicated in the past that you work with Forams

Actually, I'm vert paleo these days. I merely know folks who work with forams, and am reporting what they've told me. I've seen enough data to agree with their conclusions, but am not an expert.

Of course, but I think you take this too far. Each taxa diagnostic for each taxonomic level?

I'm not sure how this CAN be taken too far. What I mean is, in Decapoda, there are certain traits that differentiate between the various suborders and families. Within, say, Brachyura, there are DIFFERENT traits that differentiate between THOSE sub-groups. Within Raninoida, there are DIFFERENT traits that differetiate between THOSE sub-groups. All of these traits are found in each species--a Raninid has all the traits of Brachyura and Decapoda--and it takes a fair amount of expertise to differentiate between them.

Ever see the Treatise? (probably has a longer name, but in invert paleontology it's just, The Treatise.) If you have, that's the kind of thing I'm talking about. It doesn't go on for pages for each taxonomic level; it merely lists the diagnostic characters for each taxon. It's necessary given a nested hierarchy concept of taxonomy, which is in turn necessitated by the fact that life evolved. You can't grow out of your ancestry, after all.

Wait, is that what that means? I'm not sure. This is where I need to read the paper and get a better idea of what they did. My gut tells me that they should have only included in their analysis examples using ONE strictly applied species concept, rather than jumble them all together which is what it superficially appears they did.

I think we can agree that there's a gulf of difference between "we can move between them as needed without any hesitation, so long as we acknowledge the limitations of each concept" and "jumble them all together". Jumbling things all together rather indicates that one has not properly acknowledged the limitations of each concept.

You don't have any problem with them running the analysis on examples that applied different species concepts?

Depends on how they do it, but fundamentally, no. I CAN'T have such a problem. No one who works with cladograms that include both extant and extinct taxa can. When you do phylogenies that include extinct and extant species, this happens all the time. There are methods to address the issues that arise from such mixtures of methods, and obviously great caution must be exercised, but it's standard procedure anymore.

So here is an example - and I'm sure you can think of many more of these than I can - in which a morphospecies concept is limiting, just as there are limitations in applying genetic and biological species concepts.

I don't know enough about the species in question to know if the example is valid, but I suspect that there are more differences than merely the beak length. I may be wrong, but in most of the cases I've seen that's been the case. Even when working with just skulls and teeth--even when just working with teeth--there are multiple characters involved.