Hacker News new | comments | show | ask | jobs | submit login
Galapagos finches caught in act of becoming new species (bbc.com)
61 points by sohkamyung 8 months ago | hide | past | web | favorite | 55 comments

For those who are scientifically minded but not biologists, I highly recommend the Grant's "How and Why Species Multiply" [0]. It's very well written. If you're much more into population dynamics, you could give their book "Ecology and Evolution of Darwin's Finches" [1] a look. It's much more technical, being aimed at maybe upper level undergrads (in evolutionary bio) or so. They have another book, "40 Years of Evolution: Darwin's Finches on Daphne Major Island", but I haven't read that one yet. Guess I should put it on my amazon wishlist!

[0] https://www.amazon.com/How-Why-Species-Multiply-Evolutionary...

[1] https://www.amazon.com/Ecology-Evolution-Darwins-Finches-Pet...

Most interesting part for me:

"And in this paper, new genetic evidence shows that after two generations, there was complete reproductive isolation from the native birds. As a result, they are now reproductively - and genetically - isolated. So they have been breeding exclusively with each other over the years."

If you have never read "The Beak of the Finch: A Story of Evolution in Our Time" [1], do yourself a favor and get a copy asap. Not only is it a realy good read on the work of the Grants and their team, it is one of the very best pop-science books I have ever read. [1] https://www.amazon.com/Beak-Finch-Story-Evolution-Time/dp/06...

I was gobsmacked when I asked as a postdoc: "How do new species evolve" and was told: "something something physical seperation between breeding populations leading to breeding incompatibilities, but we don't really know something something"

So they couldn’t give you evidence, just assumptions?

Their is plenty of evidence for population drift. Why did native Americans have a different skin coloration than the rest of the world?

Remember, the land bridge was recent and humans with a long reproductive cycle still had very noticeable drift over ~1,000 generations. Which was not simply adaptation like reducing pigmentation when moving North.

Genetic incompatibility is actually rather difficult to understand in detail as in what specifically stops working and why that occurred, but easy to understand at a high level.

Exactly. The biggest real example I know of is ring species. Everything else seems to be total handwaving. In fact "evolution" is just an umbrella term for many different theories about how speciation happens.

I have no problem with the theory of common ancestry. That has a lot of proof. But the idea that speciation happens EXCLUSIVELY by random mutation and "natural selection" is just handwaving theorizing. It's super hard to prove and we haven't even begun to get close to what constitutes as a proof. It's on the level of ancient people thinking the ONLY way to get fire is by rubbing some sticks together.

And yet I see the latter ("natural selection") being used by many in the scientific community to suppress other theories even though it is completely handwavy and vague. That is bullying.

Basically common descent can be rigorously defined and shown. The PROCESS by which speciation happens is poorly understood and has been a subject of much posturing by the scientific community (including popularizers like Dawkins) for decades.

(PS: if you're going to knee jerk downvote this because it criticizes "evolution", at least address it with substance.)

I downvoted you (sorry, It's not often I do), and it wasn't knee-jerk. Because:

1. In both your comments on this thread you seem to want to relitigate some beef with 'random mutation and natural selection', even though that wasn't the point of, and isn't directly relevant to the article cited, Which is about the (predicted, but now for the first time observed) role of hybridisation in speciation.

2. You seem to have left your understanding of the mechanisms of evolution somewhere in the 1970s. I did my PhD in the 1990s, and I don't recognise "The idea that speciation happens EXCLUSIVELY by random mutation in natural selection" [caps yours], Except -- to be honest -- from historic papers and antievolution websites (though I'm very willing to believe it is a misunderstanding shared by journalists). Certainly the people I studied with had a much more nuanced view of evolutionary dynamics, not to mention one that didn't assume sexual reproduction or closed species.

3. Proof is for mathematicians. Obsession with proving evolution is again, something I only know from creationist rhetoric.

4. Citation needed for your bullying claim. There are certainly some who think natural selection is the most important mechanism in evolutionary change, there are those who would emphasise genetic drift, they go to the same conferences. Your image of the state of the science is fiction, in my experience. (Though I've been out of academia for 15yrs).

What is your specific objection?

Their is plenty of examples where even short term separation results in population drift: EX: Skin tone among native Americans is unique even with only ~1,000 generations separating the land bridge from modern times.

In the short term this may be adaptation to a new envornment but eventually random mutation takes over.

Their is evidence that closely related members may have difficulty breading with anything from 1% reductions in fertility to 99%. Which arguably makes them the same species as they can still in theory breed.

However, adaptation can make it's own issues. Grizzly's + Polar bears is a solid example where offspring is fertile but poorly adapted to either of their parents environment.

So, now you have examples of drops in fertility and reductions in reproductive success as populations are separated. Nothing is going to stop this process from hitting zero net fertility and thus a new species.

My objection is just that: genetic drift != speciation. They are still completely interfertile with Europeans after 1000 generstions! Yet here we have a breathless article reporting noting more than an existing male from a "separate species", which is in fact interfertile and had viable offspring, as being proof of "evolution"!

We don't know how speciation happens, and many people including scientists use a lot of unscientific "just so" stories.

To claim we know that every instance of speciation happens through 1. random mutation and 2. "natural selection", a catch-all term which is not well-defined and signifies very little, is not very scientific. But worse than that, it is then conflated with the theory of common descent, and the emergent sociological result is that many scientists and non-scientists constantly co-opt the word "evolution" to say "evolution is true and creationism is therefore false" or whatever. There have been different theories such as punctuated equilibria but we don't actually KNOW how speciation happens. It's all posturing and catch-all terms.

1,000 generations is an incredibly short period ed: genetically. Dinosaurs to people is ~20,000,000 generations depending on changes in reproductive cycles likely significantly more than that.

The shocking thing is how many differences showed up not that we are still mostly though again not 100% comparable. Remember, fertility treatments among young people are a thing and incompatibility first shows up as an increase in the natural abortion (miscarriage) rate ~(30% to 50%). You can find set's of people such that A/B and C/D are more fertile than A/D or C/B.

Right, but how do you get from that, say, to cats and dogs? Or the evolution of wings?

In the former, you need to have some sort of event that separates the populations for millions of generations or they'd constantly be interbreeding. But allopatric speciation examples are few and far between. OR, you know, there could be ANOTHER mechanism besides random mutation and "natural selection" (handwavy term) that causes speciation.

In the latter, how a morphological feature came to evolve when the intermediate steps would have most likely been valleys in the fitness function is an interesting question. Proto-wings do not seem to confer any evolutionary advantage other than the speculative one trotted out by people who must have the development of morphological features proceed ONLY by random mutation and natural selection.

No one is disputing those two things take place. The question is whether they explain ALL SPECIATION.

Many kinds of protowings are useful and currently exist.

Feathers for example provide great insulation.

Flying squirrels demonstrate another example. Important to note with flying squirrel vs bats is being even slightly better at flying is useful for squirrels. From their being slightly better at falling to flight is a steady improvement. Even normal squirrels splay their hips sideways to catch air and glide somewhat: http://feedingnature.com/how-high-can-a-squirrel-jump-off-th... Even their tails get into the act.

Flying fish similarly benefit from jumping out of the water, as part of a continuum of other fish that jump slightly further than normal out of the water etc.

Even seemingly useless stubby wings provide for threat displays like a Cobra's hood. Aka, I can make my self look bigger without needing to grow big.

Evolution is often just as much about removing things as adding. So, intermediate form may look clunky and then get cleaned up at a local optimum.

PS: Insects provide some great examples of wings with various amounts of utility.

>My objection is just that: genetic drift != speciation.

What time-scales are we talking about here? If it's on the order of hundreds or thousands of years, then yeah, you don't necessarily expect speciation to occur. If your time-scales are on the order of hundreds of thousands or millions of years that's different - and hundreds of thousands of years may be too small. Are you trying to claim that genetic drift over geological time spans does not make speciation highly likely?

>They are still completely interfertile with Europeans after 1000 generstions!

A thousand generation is nothing.

Are you trying to claim that genetic drift over geological time spans does not make speciation highly likely?

Correct! Because in every generation it is far more likely that the mutant strains would be brought "back into the fold" through interbreeding, and you need the precise opposite effect to happen over many generations to produce a separate species or morphology. So the probability decreases exponentially with generations, not increases.

This very article shows that the male finch was not a different species since it's interfertile with viable offspring.

You are ignoring the need for the mutants to reproduce and outcompete the nonmutants eventually, while somehow simultaneously making a clean break with the other species. The process has not been understood at all.

>Correct! Because in every generation it is far more likely that the mutant strains would be brought "back into the fold" through interbreeding

Is that something you came up with yourself or is it something that mainstream biology acknowledges?

And if random-mutation+selective pressure (i.e. natural selection) is not the primary mechanism for speciation, then what is?

That's an argument from authority and argument from ignorance.

I am happy to say it's pretty evident we don't know.

If we do know, I want to hear the actual substance. What do we know, HOW do we come to know it and where's the proof that mathematically it makes any sense?

Mathematically, beneficial mutation X spreads to the population A. Beneficial mutation Y that depends on Y spreads to population A.

However without 2 copies of X, Y becomes harmful. So, it only spreads when the population A already has 2 copies of A.

Now, population B does not have any copies of X and pairs can only get one copy of X from a parent, but will also get 1 copy of Y from a parent resulting in offspring that have issues.

Critically, dependency between mutations is extremely common for example the shape of the bones in your foot are related causing an ever increasing number of incompatible issues. Thus, mathematically independent populations will become less compatible over time.

PS: Mules start to show up when only one copy of X is required so getting one X and one Y is enough. However, offspring are not going to always get a copy of X.

>That's an argument from authority and argument from ignorance.

It's neither. I just asked you where you got your ideas from. And it sounds like you are dismissing an entire field of scholarship because you thought about this really really hard.

Agreed, and it's depressing to see that so many people see any disagreement with "natural selection pressure and mutation is the sole method for formation of new species" (despite these two being rather hand-wave-y in parts) as anti-scientific.

>But the idea that speciation happens EXCLUSIVELY by random mutation and "natural selection" is just handwaving theorizing.

What alternative mechanisms are there?

Who knows. But we shouldn't use an argument from ignorance to say "X did it".

The problem is with the concept of species. It's a very informal concept, introduced because we like to categorize.

So these "new" birds are distinct from all the other birds on the island that they form a separate breeding group. But how distinct are they from the outsider that started it all? Would they breed with birds from that birds home island? If so, I'd say they're not really a new species but a displaced species that survived one generation of interbreeding with the locals.

While certainly interesting, I wouldn't say this is a great example of speciation. One might even argue that it's an invasive species.

The beautiful thing about Life (and it's complexities) is how fluid and continuous (as opposed to stratified and segregated) it is.

Depending at what scale (in size and time) you observe, we are all one species.

Even in biology there are multiple definition of species. It shows how hard it is to pigeon-hole Life.

Whether or not the speciation is "complete" is not really the point - the title of the article says they are "...caught in the act of becoming new species" - in fact the article highlights the difficulty in stating what determines something to be a separate species.

I'm just saying they may well be an outside species as opposed to a new one. There is certainly no new genetic material or variation that arose.

I learned in high school that species are distinct when group B can no longer produce viable offspring with group A.

Is it more complex than that?

Yes very much so. And in fact you could already tell that that was not going to be the right answer, because also in high school, you learned that some species don't reproduce sexually.

In fact even within sexual species there are a whole range of reproductive incompatabilities: from anatomical incompatibility, to impossible fertilisation, developmental dead-ends, through to live but infertile offspring, and seemingly viable offspring with dramatically lower environmental fitness. Species are an organising model, very useful in a lot of biological reasoning, but they are not a biological absolute.

In general, whenever you were taught something black-and-white with hard boundaries in high school science, they were probably oversimplifying.

Yes. The answer to this question[1] on biology.se gives a good rundown of the issues facing this and other definitions of species.

The short version is that this definition only works for sexually reproducing species. The long version is that it doesn't work.

1: https://biology.stackexchange.com/q/39664/4101

So deniers of evolution will finally stop?

No, there are still plenty of people who sincerely believe in a flat Earth, that vaccines are a conspiracy, and that homeopathic "treatments" make them better. Best to move on and not worry about people who chose not to have evidence-based beliefs.

Homeopathic treatments help lots of people. The trick would be to demonstrate that they help people more than a placebo.


The placebo effect is to a large extent due to regression to the mean, i.e. a statistical glitch that has nothing to do with health outcomes.



<spock>Tricorder readings report fascinating regressions.</spock>

"Statistical regression to the mean predicts that patients selected for abnormalcy will, on the average, tend to improve. We argue that most improvements attributed to the placebo effect are actually instances of statistical regression."

<bones>I'm a doctor, not a faith healer!</bones>

"Thus, we urge caution in interpreting patient improvements as causal effects of our actions and should avoid the conceit of assuming that our personal presence has strong healing powers."

<kirk>You're a healer, there's a patient. That's an order.</kirk>

<scott>Now, you're an engineer!</scott>

If you don't think patient perception has anything to do with patient wellness then honestly I just hope you're not in medicine.

The problem is when they start voting based on those beliefs and help elect representatives who then attempt to make policy for all of us based on those beliefs.

Don't worry, if their beliefs are truly wrong, concrete consequences will start happening and people will see the error of their ways. For example if enough people stop vaccinating their children, common ilnesses will start spreading again and people will clamor for a fix.

My sense of the available evidence of the behavior of social groups with blatantly mistaken beliefs is that many don't 'see the error of their ways.'

When you apply this statement to other subjects (eg religion in government) it is rather depressingly incorrect.

Well theocratic governments used to be the norm in Western society. Then people realized that things like kings "ordained by God" and other such nonsense wasn't actually good for then. This might have to happen on a multi-generational level, but regardless people eventually realize what's good for them.

Lastly when you realize not to worry about people with illogical beliefs, and instead focus on educating those who are either open to information, or already support your beliefs you'll focus your energy on the things that you can actually change. Yelling, worrying or typing "you're wrong" to dogmatic religious adherents, homeopathic healers or conspiracy theorists is not going to change their mind.

I'm unwilling to be unworried by preventably sick kids who will carry the remnants of a serious illness for the rest of their lives. The consequences can be avoided if we act and talk now.

Do you think they shouldn't force their policy into you, but you can force yours into them? Why is your belief that you are right more entitled to use of force than theirs?

Policy based on science trumps bronze age mythology, so "it depends".

It is not obvious why.

To some. Yet it's still true. Go figure. If you honestly think about your bronze age superstitions enough you will eventually figure out why they can't possibly be true, and aren't even remotely ethical. Many other people, even children, have found the path out of that trap, and you can too, so don't give up hope. It will become obvious eventually.

Some people believe in interspecies evolution, but not between different species.

Correct, "of the same kind". The birds didn't become another species say "bird -> bat".

If all I had was a cell slurry I'd hardly be able to tell the difference. Where does one kind end and the other begin? Maybe the kind of earth is DNA-based with left-chiral protiens and right-chiral metabolism.

there was a time when birds and bats were hard to distinguish in the sense that both their ancestors were first one, then very closely related, and now very distinct and separate. And it's hard to tell where to draw the line in this process of separation.

Wouldn't this just add fuel to the young-Earth idea? If evolution can happen so fast, those billions of years might not be so billion after all...

So hybridization is now the canonical mechanism through which new species arise?

I thought it was autonomous mutation plus natural selection.

Maybe we have to face the fact that there's multiple ways to reach a solution and no single canonical mechanism. (essentially the point of the article)

One point is super confusing in this article, and also gives massive fodder to critics of evolution:

"We tend not to argue about what defines a species anymore, because that doesn't get you anywhere," said Prof Butlin. What he says is more interesting is understanding the role that hybridisation can have in the process of creating new species, which is why this observation of Galapagos finches is so important.

If you can't define what a species is, how can you talk about the process of creating new ones?

It seems they should be talking about the rate of gene flow between groups. "Genetically distinct" means very little because phenotypes can vary widely and still interbreed. Look at dogs, for example.

I would like to see more examples of allopatric speciation or other speciation. What they have here so far is reduced gene flow to a group of finches. Proving that none of them CAN interbreed with females is not so easy, as anecdotal evidence of females not recognizing songs of males doesn't cut it.

It's a bit like proving that an animal species has truly homosexual individuals. None of them do (except possibly for domesticated sheep) -- in fact the theory of evolution would predict that homosexuality would greatly reduce fitness, but there COULD be some members that are lifelong homosexuals, same as beta wolves etc. However to PROVE that an organism is exclusively homosexual they would need to carefuly watch it and all its pairings, however brief, its whole life.

Now, with species, you have to watch EVERY male with EVERY female around it, and vice versa. That is a huge undertaking and I doubt they did it here.

All they observed is phenotypical differences. And they haven't really explained how this is different from eg dogs, except to say that the females THEY SAW did not prefer the males' songs. Certainly this shows reduced gene flow due to preferences but not that they aren't interfertile. This is much different than a cat and a dog for example.

What I am ultimately trying to say is, I am quite skeptical that we have discovered ALL the means by which speciation occurs.

You can be a creationist or whatever else you'd like to be at this point. We barely understand how speciation really takes place, in the end we still only have theories.

I have read Lee Spetner's book "Not by Chance" which goes into the mathematics of speciation occurring exclusively by random mutation and natural selection. And it is astronomically improbable on its own terms.

This there HAVE to be forcing processes from the outside which we don't know about yet. We only guess into what they could be.

And on a related note, using evolution to explain behaviors and traits is like a religion, full of "just so" stories, in fields such as evolutionary psychology but also to explain pretty much ANY trait in a genetic fitness framework.

These "just so" stories are so pervasive yet smack of a notoriously unscientific practice and framework. All one has to do is concoct a possible way by which a trait (proto-wings, homosexuality, men becoming sleepy after an orgasm -- anything really) might confer some advantage. And the other way, such "just so" stories are used to "explain" the presence of certain traits as "the evolution of those traits" via totally untestable stories about the past.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact