Join 3,418 readers in helping fund MetaFilter (Hide)


Stick Insects Argue Against Evolution
January 16, 2003 11:47 AM   Subscribe

The Twisted Path of Stick Insect Evolution Challenges the Theory of Evolution
Sort of... In an article on the cover of today's Nature, research on the evolution of stick insects is announced. But depending on the news source, this is either a strong challenge to the theory of evolution, or a mild revision. So who's right?
posted by rschram (48 comments total)

 
Evolution is not a "theory." It happens. Darwin had a theory about how it happened, they call that Darwin's Theory of Evolution, but evolution itself is not theoretical.
posted by Pollomacho at 11:59 AM on January 16, 2003


Brief summary: A certain insect evolved to not have wings. Sometime later it re-evolved to have wings. This is evidently the first time that re-evolution of a complex ability has been reported. If anything, this is a clarification about how evolution works.

strong challenge to the theory of evolution

The articles you linked to did not portray this as a strong challenge to evolution at all.
posted by jsonic at 12:04 PM on January 16, 2003


From the first linked article:

Whiting, an assistant professor of integrative biology, and Maxwell found that members of a certain group of insects lost the ability to fly and then re-evolved it 50 million years later - a conclusion that means the theory of evolution itself must continue to change. (emphasis added)

Um, no. It means selection pressures on the insect must change over time.
posted by monju_bosatsu at 12:08 PM on January 16, 2003


Brief addition to pollomacho:

Evolution is not a theory in the sense of its common usage: a possible but not proven hypothesis.

Evolution is a theory in the scientific sense: an explanatory framework of a real and ongoing process, cf. the theory of gravity.

Also, like jsonic, I don't see how this is a strong challenge in any way. Any evolutionary scientist will tell you that the theory (remember, this means "explanatory framework") is constantly revised according to evidence, even in minute ways (but evidence is, of course, not constituted by fables).
posted by The Michael The at 12:09 PM on January 16, 2003


I can't wait to see creationists trying to play this into their ideas about intelligent (but apparently indecisive) design. Why would an all-knowing God put wings on an insect, take them off again, and then decide He liked them better before?
posted by wanderingmind at 12:10 PM on January 16, 2003


monju_bosatsu:

The change in selective pressures certainly does occur, but I think that they meant that science must revise the theory of evolution, not that it itself changes.
posted by The Michael The at 12:10 PM on January 16, 2003


Also... important to point out that terms thrown around like "re-evolution" and "de-evolution" make no sense and are predicated in misunderstanding.
posted by cadastral at 12:11 PM on January 16, 2003


Furthermore to my above post, there is no one "the" theory of evolution. You could say that the works may challenge current evolutionary theories (for sake of argument, jsonic) but not that it pesents a "strong challenge" to THE theory of evolution.

I do find it interesting that BYU did the research, I didn't know how the LDS stood on evolution, I knew they dig genetics/geneology, but evolution, I had no clue. Can anyone clarify?
posted by Pollomacho at 12:14 PM on January 16, 2003


"Ever noticed that people who believe in Creationism look really unevolved?" - Bill Hicks
posted by inpHilltr8r at 12:17 PM on January 16, 2003


We already knew that birds and insects evolved wings separately. Therefore, we knew that evolving wings isn't a once in history kind of thing. So why couldn't insects evolve them twice?

Not to make an ad hominem, but I really think you could do better than BYU for information about evolution. :) Although even their article doesn't claim that this is a strong challenge to the theory of evolution. It refers to the finding as "a conclusion that means the theory of evolution itself must continue to change. Well, duh. Until we know everything about evolution (and we never will) it will always change when new data is accumulated, just as the Theory of Gravitation changes when new discoveries about relativity or quantum theory are made. That's just how science works.
posted by callmejay at 12:20 PM on January 16, 2003


"Why would an all-knowing God put wings on an insect, take them off again, and then decide He liked them better before?"

To test your faith, of course.

Why must I continually point out the obvious?
posted by mr_crash_davis at 12:30 PM on January 16, 2003


Pollomancho: Mormonism and Evolution

Upon the fundamental doctrines of the Church we are all agreed. Our mission is to bear the message of the restored gospel to the world. Leave geology, biology, archaeology, and anthropology, no one of which has to do with the salvation of the souls of mankind, to scientific research, while we magnify our calling in the realm of the Church ... the First Presidency
posted by blue_beetle at 12:32 PM on January 16, 2003


Why would an all-knowing God put wings on an insect, take them off again, and then decide He liked them better before?

That proves it: God is a woman.
posted by timeistight at 12:34 PM on January 16, 2003


Definitely not a challenge to the theory of evolution. Not even close. If anything, it makes it stronger. The ability to lose and regain traits would seem to be nothing if not a very strong testament to selection pressure.
posted by charlesv at 12:38 PM on January 16, 2003


i'm with monju (literally: i'm sitting in class with him right now (God bless wireless)): it means the definition of the 'fittest' changes over time. hardly surprising.
posted by jjrr at 12:46 PM on January 16, 2003


Whiting, an assistant professor of integrative biology, and Maxwell found that members of a certain group of insects lost the ability to fly and then re-evolved it 50 million years later

50 million years later? Well, everyone knows God created the earth 5000 years ago, where did they get this goofy 50 million year number? It must be a typo...
posted by joecacti at 12:50 PM on January 16, 2003


Does this mean........I could re-evolve a tail?......Cool.
posted by troutfishing at 1:04 PM on January 16, 2003


Didn't various cetacean species evolve to become land-walkers and then (x amount of years) later return to the sea? This, to me, seems to suggest the same points the article marvels over as being something completely new in scientific research, though perhaps in opposite thrust.

Oh, and insert obligatory I *heart* mr_crash_davis here.
posted by WolfDaddy at 1:06 PM on January 16, 2003


That proves it: God is a woman
I guess the Devil has been spotted too.
posted by JohnR at 1:13 PM on January 16, 2003


Why would an all-knowing God put wings on an insect, take them off again, and then decide He liked them better before?

That proves it: God is a woman.


Neither of which explain why I have nipples.
posted by UncleFes at 1:38 PM on January 16, 2003


Does this mean........I could re-evolve a tail?......Cool.

Not if the chair lobby has its way.
posted by ZenMasterThis at 1:40 PM on January 16, 2003


That article presents an outright lie as a straw man to represent "evolution", that a species cannot "re-evolve a complex trait". They made up that idea out of whole cloth.

I hate it when people won't fight fair.
posted by e.e. coli at 1:58 PM on January 16, 2003


O.K. So I just finished reading the actual paper: it's pretty interesting. Those of you with access to Nature can follow this link. The authors sequenced the same region of DNA from about sixty species of stick insects and used statistical methods to reconstruct the molecular evolution of these sequences, allowing them to determine the evolutionary connections between the species. Their data reveal with a >95% probability that the initial diversification of the order must have happened in a wingless ancestral state, implying that the winged species developed wings that the ancestral species had lost in its evolution from the early winged insect ancestor. This goes against conventional entomological wisdom: it had always been assumed that if a species lost wings, genetic drift in the wingless state would make the dormant wing genes useless (since the wing genes are no longer being used, there's no selective pressure to keep them from mutating randomly over time into something nonfunctional). However, the authors point out that this conventional wisdom assumes that the genetic machinery for wings is independent from other genetic constructs, whereas in reality "it is not surprising that the basic genetic instructions for wing formation are conserved in wingless insects, because similar instructions are required to form legs, and probably other critical structures."

The punch line of the paper is as follows:
Our results support the hypothesis that the developmental pathway for wing formation evolved only once in insect diversification, but that wings evolved many times by silencing and re-expressing this pathway in different lineages during insect evolution.... To our knowledge, this is the first example of a complex feature being lost and later recovered in an evolutionary lineage, and it is possible that the reacquisition of complex features may have an important role in evolutionary diversification.

The paper says nothing beyond this about the basic mechanics of evolution, and it certainly does not "argue against evolution", as the insufferably naive title of this thread claims. In fact, the statistical analysis at the core of this paper takes the contemporary understanding of evolution as a set of fundamental assumptions, and the work does not stand without those assumptions. What this paper does provide is a new mechanism for the rapid evolution of complex traits and features.
posted by mr_roboto at 2:38 PM on January 16, 2003


That article presents an outright lie as a straw man to represent "evolution", that a species cannot "re-evolve a complex trait". They made up that idea out of whole cloth.

I hate it when people won't fight fair.


Bravo, e.e.coli. Wings have evolved at least four times, and eyes have evolved more than 40 times. It is simply untrue that a complex trait can be invented only once.

But we're not even talking about re-evolution here. Stick insects figured out a way to switch off expression of wings, and this conferred a selective advantage. 50 million years later, when they wanted them back, they found that the genes for wing expression had not deteriorated, and they could simply turn expression back on. Pretty cool, but hardly earth-shattering. This occurrence would actually be predicted by evolutionary theory: now we have a concrete example.
posted by quarantine at 3:47 PM on January 16, 2003


I must nitpick about the phrases "figured out a way to switch off expression of wings" and "wanted them back". It's all random mutations, which result in reproductive advantages or disadvantages according to the environment.
posted by timeistight at 4:11 PM on January 16, 2003


I think maybe they just misplaced them for 50 million years.
posted by Hildago at 4:22 PM on January 16, 2003


I must nitpick about the phrases "figured out a way to switch off expression of wings" and "wanted them back".

I wondered if someone would be anally retentive enough to call this out. I almost added a disclaimer to begin with. Of course they are all random mutations, but attribution of intentionality offers a concise and flippant way to discuss species-level modifications. If I say "polar bears figured out how to hide in the snow," it's obvious what I mean, right? I mean that in successive generations, the individuals that better blended into the snow tended to have more descendents, so that the species eventually converged to white color and low thermal emissions. As long as you keep real mechanisms in mind, intentionality is useful shorthand.
posted by quarantine at 4:34 PM on January 16, 2003


With something that gets misconstrued and misinterpreted as often as the theory of evolution, I think it's best to limit flights of fancy.
posted by timeistight at 4:57 PM on January 16, 2003


Note that the research was done on the genomes of 37 (the SLC Tribune's figure; I don't have access to Nature) different species of stick insects. There are "roughly 3,200 species of stick insects." 1.2% of a species does not a sample make, especially with insects, whose fecundity puts rabbits to shame (more babies = more evolution = greater diversity).

Which is not to say that I doubt the veracity of the data (If anything, this study is yet more proof for evolution, as many others have noted), it's the spin that ticks me off. Creationists will try to latch onto any science they deem to be in support of their, um, "theory." Jack Chick's hilarious interpretation of the strong (nuclear) force in the newly revised Big Daddy? Or, to not use a straw man, the constant misinterpretation of the Cambrian explosion, where the hand of god certainly must be at work, seeing as to how evolution "exploded" in the "impossibly short period" of 10 to 40 million years.

Oh, and the whole intentionality-of-organisms thing smacks of Lamarckism, which leads into another one of those funny things that creationists (and I don't mean to generalize all creationists, but one in particular I happened to cross paths with, Greg Koukl) misinterpret. He said that the communists of the 20th century killed millions of people, and were very bad men in doing so, which is certainly correct. He then said that communism is based on Darwin's theory of evolution, and therefore evolution is responsible for said genocide, because the Russians et.al. believed in the survival of the fittest, and killed off those who threatened the fitness of the state, which is complete horse-pocky. The USSR actively suppressed the study of Darwinism, seeing as to how it was derived from capitalism, and instead promoted the theories of Lysenko, which I readily pointed out during the Q&A section of the lecture--at my friend's (Christian) youth group, where I was the only (vocal) opposition. Good fun, that was.
posted by LimePi at 5:12 PM on January 16, 2003


Wow. So Jesus Christ is the strong force. Good one, Jack.

Does that make Satan the weak force?
posted by mr_roboto at 6:17 PM on January 16, 2003


I think I made this link description a little too obscure. I saw a really big difference between the portrayal in the Salt Lake City Tribune and the other coverage, including, notably, the BYU student paper. It's really something when a student newspaper has better science coverage than a mainstream profession newspaper. The results of the study are what they are, but some people twist it around to say evolution is wrong, or its arguments weakened, by the results.

The fact that the research was done at BYU doesn't necessarily get my attention. Firstly, I linked the BYU news release and the college paper because those were the most divergent from the Tribune, as well as being closest to the source of the story. BYU seems to do a lot of research—it's a very well funded institution.
posted by rschram at 6:55 PM on January 16, 2003


Note that the research was done on the genomes of 37 (the SLC Tribune's figure; I don't have access to Nature) different species of stick insects. There are "roughly 3,200 species of stick insects." 1.2% of a species does not a sample make, especially with insects, whose fecundity puts rabbits to shame (more babies = more evolution = greater diversity).

Statistical pet peeve. The quality of a sample is not measured by the percentage of the total population. In fact, there are many cases in statistics where population size is assumed to be infinite (for example, testing to see if a set of dice are "fair".)
posted by KirkJobSluder at 8:28 PM on January 16, 2003


UncleFes: Men have nipples because in animals, the female body is the template for the species (except in birds, where it's the male). It's the substitution of one Y chromosome for an X and the addition of certain hormones during pregnancy that turn the fetus into a male. (Wish I could offer a citation for you, but I forgot where I read that.)
posted by Soliloquy at 11:23 PM on January 16, 2003


Kirk-- Dice don't have babies, nor predators, nor genetic information. Stick insects do. Natural selection is best modeled by "random walks" (do I really need to post some sort of link to a Google search for you[all]? C'mon, cut and paste!), where each mutation is an independent random event. Much of today's biology (and just about every other science, for that matter) uses non-linear math to more accurately model stuff.
I find it all extremely nifty.

Aw, hell. Here's an example. Note the self-similarity of the figure, but also the crunchy jaggedness of the entire thing--the sub-triangles have the same general form, but different personalities. If I sampled 1.2% of those points, and it wasn't the right 1.2%, my results could be waaaaay off.

The amount of diversity within an insect order (which I'll assume is what the 3,200 is counting) can be absolutely mind-boggling. Apparently some stick insects are parthenogenetic (they literally fuck themselves. Heh.), while others have, um, sex. I would think that such variations in sexual behavior could easily lead to an extreme range of genes, with the inbred homogenous hillbilly parthogenites at one extreme, and the stick insect equivalent of bonobos (pygmy chimps who love to get it on) at the other.

I'm not saying that they have to research the other 98.2% of the stick insect species, and I'm sure that they tried to balance out the proportions of insects from each family (i.e., took a representative sample). Hell, I'm pretty damn sure the study's legit. It's just that I'm taking AP Biology this year, and I personally like to debate stuff a lot. And think. And stuff.

and I still can't believe that the built-in spell check has "parthenogenetic" but not the F-word in its dictionary.
posted by LimePi at 11:30 PM on January 16, 2003


Soliloquy- Yep. The Y chromosome codes for two major things, more or less--makin' sperm and the SRY gene, which makes testosterone and whatnot (some other stuff is in there, like a gene for testicular cancer, and another one for tooth enamel).
CNN has a surprisingly informative article about the history of the Y.

Men must get their X chromosome (one for each) from their mother, because Dad is the only one with the Y. Since the Y chromosome is so devoid of any really major stuff other than a schlong (which some may argue as being major stuff, but there's a thing called Turner's Syndome where the person has a single sex chromosome. Except, the person is a woman, because if the single chromosome is the Y, the zygote aborts.), that means that guys are much more like mom than dad. Girls look like both parents, though. Check out your family photo album (or someone else's, I don't care) with this in mind. Neat, eh?
posted by LimePi at 12:45 AM on January 17, 2003


Metafilter: inbred homogenous hillbilly parthogenites

Nice posts, LimePi
posted by WolfDaddy at 1:19 AM on January 17, 2003


I've been ruminating this idea over night and I was thinking, would it really be all that shocking for whales to have returned to earth? They still have the legs imbedded into their skeletal structure, they have just evolved to flippers. So in that strain, I don't know that much about stick bugs except that they always kinda creeped me out as a kid, but do the wingless ones still have wing stubs in their exo-skeletal structure? If so would it be so shocking for the wings to stop having function and then return to having function? Maybe we'll figure out that our tail starts having more function as we sit on out fat asses more and more?
posted by Pollomacho at 7:25 AM on January 17, 2003


I'm not sure what the surprise and shock is all about. I don't doubt that a lot of stick-insect DNA is "junk" DNA. That the wing-making DNA could go from useful to junk to useful again doesn't seem all that unbelieveable.
posted by five fresh fish at 9:20 AM on January 17, 2003


"it is not surprising that the basic genetic instructions for wing formation are conserved in wingless insects, because similar instructions are required to form legs, and probably other critical structures."

And they are possibly the same instructions. I've gotten the impression that genetic code is, at least somewhat, object oriented. In that case the genes could be under selection pressure not to mutate at all. What would change would be in the genes that control the less well understood area of interpretation and application in the development of the organism.

Remember, the first great virtue of computer programming is Laziness. All the evidence suggests that God is extremely (one might say Divinely) Lazy Programmer.
posted by wobh at 9:23 AM on January 17, 2003


Kirk-- Dice don't have babies, nor predators, nor genetic information. Stick insects do. Natural selection is best modeled by "random walks" (do I really need to post some sort of link to a Google search for you[all]? C'mon, cut and paste!), where each mutation is an independent random event. Much of today's biology (and just about every other science, for that matter) uses non-linear math to more accurately model stuff.

I agree, cool and really nifty but completely irrelevant to your mistake. The fact that dice don't have babies, predators and genetic information is also completely irrelevant. Statistical significance and power has nothing to do with the actual population*. It does not matter if the population size is 32 species, 32,000 species or 32,000,000 species, it has no bearing on how statistical significance or power is calculated.

So critiquing sample size by stating that it is only %1.2 of the population size is missing the point in a big way. A sample size of 37 is more than enough if you are talking about a large effect size. If you are looking for a tiny difference 370 may not be enough.


Now of course, you could critique the sample in other ways. You could question whether the sample was random or biased in some way. You could critique their insturment. There are a large number of possible critiques but without knowing the variance, we can't critique sample size.

*(Technically the population size is usually infinite because in statistics you are not comparing stick insects to stick insects but a small number of *observations* about stick insects to a theoretical distribution of all possible *observations.*)
posted by KirkJobSluder at 6:39 PM on January 17, 2003


that means that guys are much more like mom than dad. Girls look like both parents, though.

I've thought of that before & wondered where sons who look just like dad get the info if the Y chromosome is such a weak source - but there are sons who look just like dad. What about our prez, eg?
posted by mdn at 7:38 PM on January 17, 2003


I thought most everyone agreed that your prez looks like a chimp.

Oh.

Wait.

I gotcha now...
posted by five fresh fish at 8:17 PM on January 17, 2003


mdn-- There's 22 non-sex chromosomes from each parent, too. It's just that the X has a whole lot more genes than the Y does, enough to be significant, but not necessarily overpowering the other stuff.

Kirk-- I understand that national surveys and such do not have to ask everyone in America the same questions in order to fairly assess the popular opinion of all. Instead, they survey a group of people which equally represents the demographics of 290 million Americans. They can do that, because demographic information is collected with the census, every ten years.

Wouldn't getting a balanced sample of critter genomes require a sort of genetic census, where every genome would have to be analyzed in order to determine the general trend, so said trends could then be reduced into a teensy but still accurate sample?

How would they know that they have the right 1.2% without looking at the other 98.8%, is what I'm saying.

Oh, also: I have absolutely no formal or even moderately informal knowledge of stats, other than causation != correlation (sometimes). And your criticisms are totally valid and my statements were pretty vague and I have problems translating my mentalese into stuff that other people can, like, read.
posted by LimePi at 2:20 AM on January 18, 2003


Wouldn't getting a balanced sample of critter genomes require a sort of genetic census, where every genome would have to be analyzed in order to determine the general trend, so said trends could then be reduced into a teensy but still accurate sample?

No, you just need to randomly sample enough to get an idea of the variance of the population. Provided the sample is random, it does not take a large sample size to produce a reasonably accurate estimate of variance.
posted by KirkJobSluder at 8:08 AM on January 18, 2003


Gah! But how do they know if the sample is really random?!?!
It's sorta like when I would misplace my prescription eyewear and then myopically squint and walk around the house, trying to find my glasses. In order to find my glasses easily and accurately, I would need full vision--which meant, I needed to find my glasses.

So, like, how can the BYU people know where their random-sample eyeglasses are without another pair (which was my usual solution) to back them up?
posted by LimePi at 11:21 AM on January 18, 2003


P'raps you should take a first-year stats course, Lime3.14152986...
posted by five fresh fish at 7:51 PM on January 18, 2003


Yeah. Sounds like a plan to me.
posted by LimePi at 1:00 AM on January 19, 2003


fff, you got things a little mixed up there, my friend.
3.14159265358979
that's all I know - the 8 places past the part I've known forever are a nice simple pattern.
posted by mdn at 3:18 PM on January 19, 2003


« Older Muscle gals...  |  From UPI: ... Newer »


This thread has been archived and is closed to new comments