More crummy science reporting

Posted 10 July 2009 by

UPDATE 2:See my comment below for more on the study. UPDATE 1: I got a copy of the paper, and on a fast first reading things are both more complicated and more subtle than they appear from the various news stories. The paper offers a model by means of which they (purport to) essentially partition the phenotypic change in the sheep population into a portion due to phenotypic plasticity in the face of environmental change and a portion due to "evolution." "Evolution" as it is used in the news stories (and, it appears, in the paper) is construed very narrowly as an adaptive change in the genetic composition of the population due to natural selection. Somewhere Larry Moran is gnashing his teeth. :) More later, after I read the paper a couple of times (or not, if a popgen person chimes in knowledgeably). =========== I posted the gist of this as a comment on Pharyngula, but in view of the recent dustups over science reporting I thought I'd promote it to a brief post here. The comment was on PZ's report of Ben Goldacre's analysis of some really lousy science reporting in the Telegraph. And it's truly crummy reporting, distorted and sensationalized. But there are also the casual errors, little misrepresentations that slip into stories almost unconsciously. For example, in the Guardian's recent story about 'shrinking' sheep there's this little jewel:
The case involves a rare herd of wild sheep on the remote Scottish island - known in Scottish Gaelic as Hirta - that are refusing to bow to conventional evolutionary pressure, which says big is best.
Evolutionary pressure says big is best? Gaaah! And the reporter has a Ph.D. in chemical engineering and is the Guardian's environment correspondent, having previously worked for Nature. Lovely.

67 Comments

Stanton · 10 July 2009

Either the reporter is an idiot who never heard of the phenomenon of "island dwarfism," or he's lobotomized himself as a sacrifice for snagging readers with trashy sensationalism.

RBH · 10 July 2009

Stanton said: Either the reporter is an idiot who never heard of the phenomenon of "island dwarfism," or he's lobotomized himself as a sacrifice for snagging readers with trashy sensationalism.
To be fair, this is apparently not a case of island dwarfism, but rather appears to be due to an easing of selective pressure associated with climate change. In that restricted sense and in that particular selective context the sentence might be interpreted to be correct, but as it's written -- "conventional evolutionary pressure" -- it's a universal claim that's false.

jasonmitchell · 10 July 2009

'bigger is better' was never the "conventional evoultionary pressure"
"selection favors those who produce more viable offspring who survive to the age of reproduction" is conventional evolutionary pressure - now I cannot tell from the article if the sheep are smaller or just have less body fat -but it seems reasonable that if winters are milder, food will be available wearlier in the spring and sheep would need less body fat to survive the winter - would be a PREDICTION of 'conventional evolutionary pressure"

here's another prediction - fecundity may change due to milder winters/ earlier spring as well

Iason Ouabache · 10 July 2009

Stanton said: Either the reporter is an idiot who never heard of the phenomenon of "island dwarfism," or he's lobotomized himself as a sacrifice for snagging readers with trashy sensationalism.
As PZ pointed out in one of his other blog posts, it might not be the reporters fault. Too often science reporters hand their work off to an incompetent editor who adds a very misleading headline, cuts important information out and reverses the meaning of whole sentences. All in the name of making a "better" story. We need to start holding editors accountable for things like this.

Corey S. · 10 July 2009

Clearly, his only experience with animal life is with Americans (especially from the midwest), who constantly seem to be bowing to 'conventional evolutionary pressure.'

(I say this as a Wisconsinite.)

stevaroni · 10 July 2009

Time magazine screwed this up too, with a similar article in last weeks (July 3rd) issue, The Incredible Shrinking Sheep of Scotland, recaping the same Science paper. It opens with this jaw dropper...

News alert: the sheep of Scotland are shrinking! On Soay Island, off the western coast of Scotland, wild sheep are apparently defying the theory of evolution and progressively getting smaller. Why? In short, because of climate change. Generally, the sheep's life cycle goes like this: they fatten up on grass during the fertile, sunny summer; then the harsh winter comes, the grass disappears and the smallest, scrawniest sheep die off, while their bigger cousins survive. That's how you end up with big sheep, which — according to Darwin's laws of natural selection — will pass on their big genes to the next generation.

Seems weird that two different news organizations would make the same mistake. I wonder if there was a press release from Science that tried too hard to sell a sexy evolution angle and ended up muddying the waters.

Henry J · 10 July 2009

So size does matter! Or as W. Rosenberg once put it, "I guess it's true - scythe matters".

RBH · 10 July 2009

stevaroni said: Time magazine screwed this up too, with a similar article in last weeks (July 3rd) issue, The Incredible Shrinking Sheep of Scotland, recaping the same Science paper. It opens with this jaw dropper... [SNIP] Seems weird that two different news organizations would make the same mistake. I wonder if there was a press release from Science that tried too hard to sell a sexy evolution angle and ended up muddying the waters.
It appears that the story came out at the same time as a major conference on science reporting in Great Britain, and it's possible the herd instinct took over. Scanning through some of the articles linked from that site, I see Kenneth Chang of the NYTimes got taken in too:
That was somewhat surprising, as larger sheep have better odds of surviving and evolution tends to favor those that are stronger.
Another offender in the half dozen stories I scanned was the Telegraph, which said
Survival of the fittest and natural selection usually means that species grow bigger as they evolve but milder weather on the uninhabited islands of the Scottish Outer Hebrides has pushed this process into reverse.
So I suspect that there was a common ancestor for the several instances, possibly the Science press release which I have not yet found online. However, AAASNow Daily News also has some weird verbiage about the phenomenon:
Call it the case of the shrinking sheep. On the remote Scottish island of Hirta, sheep have been getting smaller, shrinking an average of 5% over the last 24 years. Don't blame evolution, though. Researchers say climate change is the real culprit. ... In the past, Hirta's sheep gorged on grass during their first summer, the team notes, piling on the weight in order to make it through the island's typically harsh winters. But over the past quarter-century, Hirta has had unusually short and mild winters. As a result, Ozgul and colleagues propose, grass has become available for more months of the year, meaning the Soay sheep do not have to bulk up as much. In addition, Hirta's harsh winters used to kill small ewes born to young mothers. But now these small ewes survive--and because of their low birth weight, they never get as big as normal sheep. That drives down the average size of the entire population, the team reports. Further mathematical modeling allowed the researchers to propose that natural selection has played little--if any--role in the shrinkage of the Hirta sheep.
After reading how a change in selective circumstances led to relaxed selection that produced the decrease in mean population weight, we learn that natural selection played little role! Hm. I guess what's meant is that lower weight was not itself positively selected for, but rather that higher weight was no longer as selectively advantageous; there is relaxed selection against lower weight -- it's less disadvantageous now. But given the consistency of the stories, I suspect some not very thoughtful comments at the press conference in London at which the paper was announced (see the first link above).

RBH · 10 July 2009

On the other hand, there's a clue in the Science News:
That ability to parse effects makes the sheep study especially interesting, says evolutionary geneticist Mark Rausher of Duke University in Durham, N.C. Biologists have been working on ways to distinguish genetic from environmental effects for decades, often by comparing the specific form, or phenotype, of individuals of the same species at different altitudes. “What’s nice about this new study is that it may be the first that convincingly shows that a change in phenotype over time is driven by the direct action of the environment on phenotype,” Rausher says. More studies using this type of model could follow since “the method is very, very general,” Coulson says. He’s even considering applying it to people, whose body size is definitely not shrinking.
Guess I'm going to have to read the paper.

Henry J · 10 July 2009

Well, sheep are covered with wool, and wool can shrink when washed, so maybe these sheep were bathing more often than is typical for sheep?

Rilke's Granddaughter · 10 July 2009

Henry, you just don't have anything intelligent or entertaining to say, do you. Poor boy.
Henry J said: Well, sheep are covered with wool, and wool can shrink when washed, so maybe these sheep were bathing more often than is typical for sheep?

John Harshman · 10 July 2009

The case involves a rare herd of wild sheep on the remote Scottish island - known in Scottish Gaelic as Hirta - that are refusing to bow to conventional evolutionary pressure, which says big is best.
I wonder if this is intended to be an echo of Cope's Rule.

RBH · 11 July 2009

OK, this is a fast gloss of the original study. The authors analyzed the dynamics of 24 years of size changes in the sheep, and decomposed the variance into components employing a version of Price's equation for change in the mean value of a trait. A subsequent refinement allowed separating that change into two components, survival and fertility:
Survival-related terms are (i) changes in demographic structure caused by age-specific survival rates (termed DCs) (14), (ii) age-specific viability selection differentials, which describe change resulting from differential survival associated with the trait (VS) (15), and (iii) age-specific trait development, which describes how the average trait value changes among surviving individuals as they age (GR). Reproduction-related terms are (i) changes in demographic structure caused by age-specific reproduction (DCr) (14), (ii) age-specific fertility selection differentials which describe the difference in mean trait value between selected parents and the unselected population (FS) (16), (iii) the mean age-specific difference between offspring and parental trait values (OMD) (12), and (iv) the covariation between litter size and the difference between offspring and parental trait values (ODC) (13).
Using multiple age cohorts and data gathered over 24 years, they assessed the relative influence of the several variables on body weight. GR, DCr, and DCs contributed positively (increasing body weight) over time, while VS and FS contributed very little. OMD was the major negative contributor, decreasing body weight, and out-weighed the sum of the positive contributors yielding a net slightly negative weight change over time. They did additional analyses parsing out the sub-variables generating the dominance of OMD in reducing weight over time, finding
We next analyzed individual growth rates within each age class (7). The growth of lambs was significantly negatively influenced by August body weight and population density operating additively and via an interaction with the preceding winter’s North Atlantic Oscillation index (23) – lambs grew more slowly in years of high density following a bad winter: growth slows as competition for food increases and the amount of stored reserves required to stay alive also increases.
This is the tie-in to 'ecological' variables: the North Atlantic Oscillation is associated with warmer times in the study area, and during the 1990s (in the middle of the period analyzed in this study) the index was hitting 150-year highs. So, it nets out that the decreasing body weight is due to younger females having lower-weight lambs whose growth rates are slower on account of both being smaller and because of density-dependent competition for food to last through the next winter after a bad winter due to a climate variable. This would be a complicated study to get across to a lay audience, and since my headline, slightly exaggerated, says "crummy," it's incumbent on me to offer a better take on it. Let me make a try at a sentence or two that more accurately capture the study.
Scientists today reported a study showing how environmental effects interact with natural selection to produce change in the size of sheep. While natural selection tends to favor larger sizes of Soaey sheep on the remote Scottish island of Hirta, recent climate changes have produced smaller sheep as the longer warmer summers allow smaller females to breed successfully and smaller offspring to survive to breed themselves. [blah blah blah] So the research shows how the interaction of natural selection and environmental changes can produce what seem to be counter-intuitive changes in phenotypes.
Now, it's not clear to me that any of this is not evolution. It seems clear that the changed length and warmth of the summers eased selective pressure for larger sheep, permitting smaller sheep to successfully reproduce. Since the authors don't report genetic information, we don't know the nature of any change in the genetic structure of the population, but since body weight is highly heritable it seems obvious that it would have occurred. So the message in the research for me is that what happens to phenotypes is the net of the several variables operating in different ways to generate changes. It doesn't make sense to me to say that "evolution" refers only to adaptive phenotypic changes that are due to directional selection for pure survival due to one environmental variable. In particular, I'm not at all clear on what it means to classify a "reproduction-related term" (OMD) as not evolutionary, as the authors did. It's that reproduction-related variable that generated the size decrease.

kakapo · 11 July 2009

Rilke's Granddaughter said: Henry, you just don't have anything intelligent or entertaining to say, do you. Poor boy.
I think you're confusing Henry J and henry. I thought Henry J's comment was a funny yarn and that you're being crewel.

Bill Bigge · 11 July 2009

big is best - that explains why the only insects around these days are the size of cows and why bacteria don't exist!

I agree though,lots of little bits of bad science reporting keep appearing and it can be frustrating, I sometimes try adding comments but not every story has a comments section.

Bill Bigge · 11 July 2009

Hang on, I think I've got it, these sheep are evolving to the nano scale in order to remain compatible with modern server farms.

John Kwok · 11 July 2009

John - I immediately thought of Cope's Rule, but there is a rather obvious problem:
John Harshman said:
The case involves a rare herd of wild sheep on the remote Scottish island - known in Scottish Gaelic as Hirta - that are refusing to bow to conventional evolutionary pressure, which says big is best.
I wonder if this is intended to be an echo of Cope's Rule.
It is well documented that on islands there tends to be a tendency for metazoans to grow smaller, NOT LARGER with time, which, of course is what Cope's Rule would predict. This is especially borne out by paleontological studies of Mediterranean islands, the discovery of "dwarf" mammoths off the coast of Alaska, and perhaps, most notably, of the pygmy hominid dubbed a "hobbit" in the East Indies.

henry · 11 July 2009

stevaroni said: Time magazine screwed this up too, with a similar article in last weeks (July 3rd) issue, The Incredible Shrinking Sheep of Scotland, recaping the same Science paper. It opens with this jaw dropper...

News alert: the sheep of Scotland are shrinking! On Soay Island, off the western coast of Scotland, wild sheep are apparently defying the theory of evolution and progressively getting smaller. Why? In short, because of climate change. Generally, the sheep's life cycle goes like this: they fatten up on grass during the fertile, sunny summer; then the harsh winter comes, the grass disappears and the smallest, scrawniest sheep die off, while their bigger cousins survive. That's how you end up with big sheep, which — according to Darwin's laws of natural selection — will pass on their big genes to the next generation.

Seems weird that two different news organizations would make the same mistake. I wonder if there was a press release from Science that tried too hard to sell a sexy evolution angle and ended up muddying the waters.
It looks like cooler temperatures will be returning, thus bigger sheep will be returning. Will Solar Inactivity Lead to Global Cooling? by Larry Vardiman, Ph.D.* … Svensmark essentially theorized that when the sun is active, it deflects cosmic radiation from outer space, which reduces low cloud cover and allows more solar radiation to strike the earth’s surface, resulting in warming. When the sun is quiet, it allows cosmic radiation to increase the cloud cover, which reflects more solar radiation, cooling the earth. The sun has most recently been showing a decrease in activity, which within the next few years may well result in a decrease in global temperatures. … Svensmark, H. et al. 2007. Experimental evidence for the role of ions in particle nucleation under atmospheric conditions. Proceedings of the Royal Society A. 463 (2078): 385-396. * Dr. Vardiman is Chair of the Department of Astro/Geophysics at the Institute for Creation Research. Cite this article: Vardiman, L. 2009. Will Solar Inactivity Lead to Global Cooling? Acts & Facts. 38 (7): 12.

DS · 11 July 2009

Henry,

Cut the act and stick with the facts.

stevaroni · 11 July 2009

John Kwok writes... It is well documented that on islands there tends to be a tendency for metazoans to grow smaller, NOT LARGER

Aren't all land masses ultimately islands?

John Kwok · 11 July 2009

Yes and no. Cope's Rule holds up on continents (which I suppose one could contend is an island in the case of Australia, but even there the fossil record has a lot of evidence showing that Cope's Rule was "followed" by metazoans until human colonization approximately 40,000 years ago, when much of the megafauna became extinct.). On the other hand, there is an abundant scientific literature showing dwarfism - especially in depauperate faunas - for both living and extinct metazoans on islands:
stevaroni said:

John Kwok writes... It is well documented that on islands there tends to be a tendency for metazoans to grow smaller, NOT LARGER

Aren't all land masses ultimately islands?

fnxtr · 11 July 2009

* Dr. Vardiman is Chair of the Department of Making Shit Up at the Institute for Creation Research.
Fixed.

Monado · 11 July 2009

I spoke with a science reporter at the 2008 Science Blogging conference. She explained that editorial management insists on a "hook" that connects any scientific discovery to a benefit for humans and that the time or space allowed for describing each discovery is constantly being squeezed down, leaving barely enough conceptual leeway for a coherent sound bite. (That's not a direct quote but it's the gist of it.)

Unfortunately, science today might be in the position of women's issues in the 70s. I used to get indignant that stories of equal pay, discrimination, access to contraception, marriage laws, and the rest appeared only in the "women's pages" along with recipes and fashions. Eventually someone explained that the main editorial focus ignored those stories and the only way for them to be published at all was for the "women's editor" to give them room. The women's pages editors were basically using their limited authority to get the articles into the paper for us.

At the least, Science writers are making the public more aware of science and its effect on our lives. If we buy, read, comment, promote and suggest corrections and refinements, instead of carping, the writers might earn enough space and prestige to provide more accurate and subtle interpretations.

Monado · 11 July 2009

I spoke with a science reporter at the 2008 Science Blogging conference. She explained that editorial management insists on a "hook" that connects any scientific discovery to a benefit for humans and that the time or space allowed for describing each discovery is constantly being squeezed down, leaving barely enough conceptual leeway for a coherent sound bite. (That's not a direct quote but it's the gist of it.)

Unfortunately, science today might be in the position of women's issues in the 70s. I used to get indignant that stories of equal pay, discrimination, access to contraception, marriage laws, and the rest appeared only in the "women's pages" along with recipes and fashions. Eventually someone explained that the main editorial focus ignored those stories and the only way for them to be published at all was for the "women's editor" to give them room. The women's pages editors were basically using their limited authority to get the articles into the paper for us.

At the least, Science writers are making the public more aware of science and its effect on our lives. If we buy, read, comment, promote and suggest corrections and refinements, instead of carping, the writers might earn enough space and prestige to provide more accurate and subtle interpretations.

Monado · 11 July 2009

I spoke with a science reporter at the 2008 Science Blogging conference. She explained that editorial management insists on a "hook" that connects any scientific discovery to a benefit for humans and that the time or space allowed for describing each discovery is constantly being squeezed down, leaving barely enough conceptual leeway for a coherent sound bite. (That's not a direct quote but it's the gist of it.)

Unfortunately, science today might be in the position of women's issues in the 70s. I used to get indignant that stories of equal pay, discrimination, access to contraception, marriage laws, and the rest appeared only in the "women's pages" along with recipes and fashions. Eventually someone explained that the main editorial focus ignored those stories and the only way for them to be published at all was for the "women's editor" to give them room. The women's pages editors were basically using their limited authority to get the articles into the paper for us.

At the least, Science writers are making the public more aware of science and its effect on our lives. If we buy, read, comment, promote, and suggest corrections and refinements, instead of carping, the writers might earn enough space and prestige to provide more accurate and subtle interpretations.

Toidel Mahoney · 12 July 2009

Bill Bigge said: big is best - that explains why the only insects around these days are the size of cows and why bacteria don't exist! I agree though,lots of little bits of bad science reporting keep appearing and it can be frustrating, I sometimes try adding comments but not every story has a comments section.
Yes, it seems evolutionists like PZ Myers don't like that evolution might make predictions that could be falsified. Since evolutionism teaches that complexity increases through domination, and hence the bigger and stronger who are most successful at killing ultimately win. If the opposite sometimes happens, why can't we consider something other than Darwinism is going on? That question answers itself.

John Harshman · 12 July 2009

John Kwok said: John - I immediately thought of Cope's Rule, but there is a rather obvious problem: It is well documented that on islands there tends to be a tendency for metazoans to grow smaller, NOT LARGER with time, which, of course is what Cope's Rule would predict. This is especially borne out by paleontological studies of Mediterranean islands, the discovery of "dwarf" mammoths off the coast of Alaska, and perhaps, most notably, of the pygmy hominid dubbed a "hobbit" in the East Indies.
Well, yes, nobody is saying that Cope's Rule is true, just that its existence as a widespread idea provides partial support for the story's claim about evolutionary orthodoxy. The fact is that, even if Cope's Rule is a real phenomenon, it has a great many known exceptions, long before any Scottish sheep. So the reporter was wrong, but perhaps not completely so. And your claim is wrong too. There's a tendency for large mammals to get smaller on islands. But many other sorts can get much larger. Consider terrestrial birds (e.g. moa nalos, mihirungs, moas, elephant birds), coconut crabs, komodo dragons, etc.

John Harshman · 12 July 2009

Toidel Mahoney said:
Yes, it seems evolutionists like PZ Myers don't like that evolution might make predictions that could be falsified. Since evolutionism teaches that complexity increases through domination, and hence the bigger and stronger who are most successful at killing ultimately win. If the opposite sometimes happens, why can't we consider something other than Darwinism is going on? That question answers itself.
Nobody will recognize your characterization of "Darwinism" as anything other than a strawman. Sometimes bigger is better; sometimes smaller is better; sometimes there is a balance between advantages of both. Depends on the environment, and we can predict what will be advantageous in which environments. Your notion that natural selection is only about "nature, red in tooth and claw" is hopelessly naive. Different strategies can be advantageous to individuals, including cooperation. And there are many other factors influencing optimal size than dominance contests.

Stanton · 12 July 2009

Toidel Mahoney said:
Bill Bigge said: big is best - that explains why the only insects around these days are the size of cows and why bacteria don't exist! I agree though,lots of little bits of bad science reporting keep appearing and it can be frustrating, I sometimes try adding comments but not every story has a comments section.
Yes, it seems evolutionists like PZ Myers don't like that evolution might make predictions that could be falsified. Since evolutionism teaches that complexity increases through domination, and hence the bigger and stronger who are most successful at killing ultimately win. If the opposite sometimes happens, why can't we consider something other than Darwinism is going on? That question answers itself.
Please explain how Creationism explains better why sheep would get smaller on an island. Please explain how the disobedience of our legendary ancestors 6000 years specifically ties into this phenomenom. I've noticed that you never attempt to demonstrate why we should regard Creationism as the better alternative. So, rather than engage in your crass bigotry and grotesquely inaccurate strawmen of "Darwinism" (sic), why don't you attempt to demonstrate how Creationism is the better explanation? Oh, wait, you can't because you're a bigoted idiot.

Eric Bloodaxe · 12 July 2009

I thought that there was a trend for mammals to get smaller when isolated on islands? I have heard of pygmy elephants on some islands,(? the med) and the newly discovered fossils of small humans on an island in the South Seas.

Stanton · 12 July 2009

Eric Bloodaxe said: I thought that there was a trend for mammals to get smaller when isolated on islands? I have heard of pygmy elephants on some islands,(? the med) and the newly discovered fossils of small humans on an island in the South Seas.
It depends on the situation, actually. There have been both giant and dwarfed animals on Mediterranean islands, including pygmy elephants, pygmy Irish elk, pygmy hippopotami, giant otters, giant rodents, eagle-sized barn owls, cat-sized predatory hedgehogs... On an island, if resources are scarce, then smaller adult individuals are more likely to survive, if not prosper, because they have lower nutritional needs, while larger individuals are less likely to thrive because they would expend more energy than can be replaced with the local food supplies, especially if they expended more energy whilst competing for food. On an island, you're not going to thrive by being as big as a house if the food source is equivalent to a 10 foot by 10 foot patch of wild celery. On the other hand, there are a lot of factors that would promote the evolution of small animals into comparatively giant forms on islands, including relatively bountiful food supplies (what is a famine to an elephant can be a banquet to a hutch of rabbits), the ability to better digest a local food source, better mobility, less likely for larger, heavier individuals to be blown out to sea, lack of predators, or local predators preying on smaller individuals, etc.

Dave Luckett · 12 July 2009

I suspect that another factor might be whether individuals of the species compete for territory among themselves or not. Increased size would be less of an advantage, in that case, and might very well be overcome by the advantage of being better able to survive lean times if smaller.

stevaroni · 12 July 2009

Toidel trolls..... Since evolutionism teaches that complexity increases through domination, and hence the bigger and stronger who are most successful at killing ultimately win.

Um, no. Evolution actually teaches "make love, not war". Evolution doesn't favor "the most successful at killing", it favors the most successful at f*^%ing. The key factor to passing your genes on to posterity is ability to produce viable offspring. Now, since successfully breeding often means competing for resources and sex, and avoiding getting eaten in the proscess, successful individuals do often tend to be stronger, sleeker and healthier than evolutionary losers, but this is not always the case. For instance, there are probably 1000 chihuahuas on the planet for every gray wolf. Chihuahuas exploit a niche, fashion accessory, for which gray wolves would be badly suited indeed. Still, I'd give the gray wolf the benefit of the doubt in a cage match. From evolution's point of view, the biggest sin isn't being slow or weak, it's being unable to get laid.

If the opposite sometimes happens, why can’t we consider something other than Darwinism is going on? That question answers itself

Fine. What exactly do you propose is going on, and how do I test it to see if this is the case? It's a simple enough question Toidel, but you never answer it. What is your alternative explanation, exactly, and more importantly how do I test it?

John Kwok · 12 July 2009

Stanton, Those examples of "gigantism" apply only to small mammals and occasionally to birds like the now extinct Moa from New Zealand. But, for the most part, depauperate (By depauperate I am referring to those islands whose species diversity is relatively impoverished compared to nearby islands or the nearest mainland(s).) island faunas tend to show a trend towards dwarfism, as both the recent discoveries of pygmy mammoths (I think this was on Wrangell Island, off the coast of Alaska) and of the "hobbit" hominoid from Flores in the East Indies clearly illustrate:
Stanton said:
Eric Bloodaxe said: I thought that there was a trend for mammals to get smaller when isolated on islands? I have heard of pygmy elephants on some islands,(? the med) and the newly discovered fossils of small humans on an island in the South Seas.
It depends on the situation, actually. There have been both giant and dwarfed animals on Mediterranean islands, including pygmy elephants, pygmy Irish elk, pygmy hippopotami, giant otters, giant rodents, eagle-sized barn owls, cat-sized predatory hedgehogs... On an island, if resources are scarce, then smaller adult individuals are more likely to survive, if not prosper, because they have lower nutritional needs, while larger individuals are less likely to thrive because they would expend more energy than can be replaced with the local food supplies, especially if they expended more energy whilst competing for food. On an island, you're not going to thrive by being as big as a house if the food source is equivalent to a 10 foot by 10 foot patch of wild celery. On the other hand, there are a lot of factors that would promote the evolution of small animals into comparatively giant forms on islands, including relatively bountiful food supplies (what is a famine to an elephant can be a banquet to a hutch of rabbits), the ability to better digest a local food source, better mobility, less likely for larger, heavier individuals to be blown out to sea, lack of predators, or local predators preying on smaller individuals, etc.
Regards, John

John Kwok · 12 July 2009

John, I missed these comments of yours. Please see what I wrote in reply to Stanton's observations. I still stand by my observation that dwarfism is a common characteristic of depauperate island faunas for both living and fossil taxa. Why this is so I can only speculate that this may have to do to limited availability of resources (e. g. food, shelter) on these islands:
John Harshman said:
John Kwok said: John - I immediately thought of Cope's Rule, but there is a rather obvious problem: It is well documented that on islands there tends to be a tendency for metazoans to grow smaller, NOT LARGER with time, which, of course is what Cope's Rule would predict. This is especially borne out by paleontological studies of Mediterranean islands, the discovery of "dwarf" mammoths off the coast of Alaska, and perhaps, most notably, of the pygmy hominid dubbed a "hobbit" in the East Indies.
Well, yes, nobody is saying that Cope's Rule is true, just that its existence as a widespread idea provides partial support for the story's claim about evolutionary orthodoxy. The fact is that, even if Cope's Rule is a real phenomenon, it has a great many known exceptions, long before any Scottish sheep. So the reporter was wrong, but perhaps not completely so. And your claim is wrong too. There's a tendency for large mammals to get smaller on islands. But many other sorts can get much larger. Consider terrestrial birds (e.g. moa nalos, mihirungs, moas, elephant birds), coconut crabs, komodo dragons, etc.

Stanton · 12 July 2009

Of course: I didn't say that gigantism is the only route for small species that find themselves on islands. In some cases, yes, the species will get smaller. The ones that I can think of off the top of my head are some of the Pleistocene Sardinian otters, which were smaller than the European river otters. On the other hand, in some situations, a founding population, or a group of related founding populations may go in opposite directions: going back to Pleistocene Sardinia, besides the small otters, we also find the giant, human-sized clawless otter, Megalenhydris, too. And going back to New Zealand, many people don't realize that the moa ranged in size from turkey-like fowl to monsters that made ostriches look comparatively scrawny.

John Kwok · 12 July 2009

Stanton, Thanks for your remarks, with which I do concur, but I also hope you realize that it's really those islands which have depauperate or "impoverished" faunas where one tends to see dwarfism:
Stanton said: Of course: I didn't say that gigantism is the only route for small species that find themselves on islands. In some cases, yes, the species will get smaller. The ones that I can think of off the top of my head are some of the Pleistocene Sardinian otters, which were smaller than the European river otters. On the other hand, in some situations, a founding population, or a group of related founding populations may go in opposite directions: going back to Pleistocene Sardinia, besides the small otters, we also find the giant, human-sized clawless otter, Megalenhydris, too. And going back to New Zealand, many people don't realize that the moa ranged in size from turkey-like fowl to monsters that made ostriches look comparatively scrawny.

John Harshman · 13 July 2009

John Kwok said: John, I missed these comments of yours. Please see what I wrote in reply to Stanton's observations. I still stand by my observation that dwarfism is a common characteristic of depauperate island faunas for both living and fossil taxa. Why this is so I can only speculate that this may have to do to limited availability of resources (e. g. food, shelter) on these islands:
It's a common characteristic of *large mammals*. Pretty much limited to that, as far as I know. If you have examples of island dwarfism that do not involve large mammals, I would be interested. What I most object to is that you generalize this into a claim about metazoans. Are island nematodes dwarfish? Doubtful. Island insects? No. Island reptiles? No, and in fact there seems a more likely trend toward gigantism (e.g. Galapagos and Aldabra tortoises, the afore-mentioned Komodo Dragon, etc.). It seems to me fairly reasonable that island dinosaurs (plesiomorphic ones, that is) might have shown a trend similar to that of large mammals, but I don't know of any relevant fossil record. In your statement above, if you substitutes "large mammals" for "fauna", I would have no argument with you.

John Kwok · 13 July 2009

John,

Please bear in mind what I wrote with regards to "impoverished" faunas. The Galapagos Islands have high species diversities due to adaptive radiations within several bird lineages, not only the Galapagos Finches but also mockingbirds. But if you want to substitute "mammals" for metazoans that is fine with me.

Regards,

John

John Harshman · 13 July 2009

John Kwok said: John, Please bear in mind what I wrote with regards to "impoverished" faunas. The Galapagos Islands have high species diversities due to adaptive radiations within several bird lineages, not only the Galapagos Finches but also mockingbirds. But if you want to substitute "mammals" for metazoans that is fine with me. Regards, John
This is nonsense. The Galapagos have quite low species diversity. The fact that a small radiation of tanagers ("Darwin's finches") and a few mockingbirds make up most of the passerine fauna does nothing to change this. Certainly Sicily has a much more diverse fauna. (And there is only one adaptive radiations of birds in the Galapagos, not "several"; further, the mockingbirds can't be called an adaptive radiation since the species are all allopatric and all occupy more or less the same niche.) You need to stop this line before you tie yourself in further knots. But I will ask again: if you know of any island dwarfism other than in large mammals, please mention it now.

Kevin B · 13 July 2009

Would it be reasonable to hypothesise that, in the absence of other factors (eg competition, predation,) a "medium-sized" mammal has an advantage over either the very large or the very small?

John Harshman · 13 July 2009

Kevin B said: Would it be reasonable to hypothesise that, in the absence of other factors (eg competition, predation,) a "medium-sized" mammal has an advantage over either the very large or the very small?
No. In the absence of other factors, nothing has an advantage over anything else. The question is what combination of many possible factors exists in a particular population's environment. There may be an optimal body size for males, and another for females. There may in fact be more than one optimal body size for males (dominant male vs. SLF, for example).

John Kwok · 13 July 2009

If I'm not mistaken, the Flores Island hominid fossil, known colloquially as the "hobbit", was discovered along with a dwarf version of the Komodo dragon. But again, dwarfism is largely endemic to small islands noted for their impoverished faunas, and yes predominantly with mammals known to have larger relatives on species rich mainlands.. As for the Galapagos Islands, the Galapagos Finches apparently also evolved primarily via allopatric speciation, along with the mockingbirds:
John Harshman said:
John Kwok said: John, Please bear in mind what I wrote with regards to "impoverished" faunas. The Galapagos Islands have high species diversities due to adaptive radiations within several bird lineages, not only the Galapagos Finches but also mockingbirds. But if you want to substitute "mammals" for metazoans that is fine with me. Regards, John
This is nonsense. The Galapagos have quite low species diversity. The fact that a small radiation of tanagers ("Darwin's finches") and a few mockingbirds make up most of the passerine fauna does nothing to change this. Certainly Sicily has a much more diverse fauna. (And there is only one adaptive radiations of birds in the Galapagos, not "several"; further, the mockingbirds can't be called an adaptive radiation since the species are all allopatric and all occupy more or less the same niche.) You need to stop this line before you tie yourself in further knots. But I will ask again: if you know of any island dwarfism other than in large mammals, please mention it now.

John Harshman · 13 July 2009

Please stop.

The Komodo Dragon is a giant monitor lizard. Any monitor lizard found on Flores isn't a dwarf form of the Komodo Dragon unless you want to postulate that it had a giant ancestor, and I would be interested in the reasoning for that.

The mode of speciation of Darwin's finches or Galapagos mockingbirds isn't relevant to whether they're adaptive radiations. Adaptive radiations involve the invasion of multiple niches. Darwin's finches, check; mockingbirds, no.

At least we now agree that island dwarfism appears to be limited to large mammals.

John Kwok · 13 July 2009

John - I believe I saw an abstract of a paper on the Flores hominid in which it was believed that it had used stone tools associated with it to hunt a pygmy version of the Komodo Dragon. As for Darwin's finches, they did indeed radiate in such a way that involved "the invasion of multiple niches". If you want to really be happy, I can point to such extensive adaptive radiations as those of cichlid fishes in the East African rift lakes as classic examples:
John Harshman said: Please stop. The Komodo Dragon is a giant monitor lizard. Any monitor lizard found on Flores isn't a dwarf form of the Komodo Dragon unless you want to postulate that it had a giant ancestor, and I would be interested in the reasoning for that. The mode of speciation of Darwin's finches or Galapagos mockingbirds isn't relevant to whether they're adaptive radiations. Adaptive radiations involve the invasion of multiple niches. Darwin's finches, check; mockingbirds, no. At least we now agree that island dwarfism appears to be limited to large mammals.

Stanton · 13 July 2009

If the giant monitor lizards contemporaries of Flores Man were smaller ancestors of the Komodo dragons, then it would be more appropriate to refer to them as "a smaller ancestral form of Komodo dragons," as referring to them as "pygmies" would imply that they were even smaller than typical monitor lizards.

Or, it could be that the monitor lizard contemporary may have been the extinct species Varanus hooijeri, which is a relative of the Gray's Monitor, V. olivaceus (and as such, referring to it as a "pygmy Komodo dragon" would be totally inappropriate).

John Harshman · 13 July 2009

John Kwok said: I believe I saw an abstract of a paper on the Flores hominid in which it was believed that it had used stone tools associated with it to hunt a pygmy version of the Komodo Dragon. As for Darwin's finches, they did indeed radiate in such a way that involved "the invasion of multiple niches". If you want to really be happy, I can point to such extensive adaptive radiations as those of cichlid fishes in the East African rift lakes as classic examples:
No, really. Please stop. At least read what I write. Yes, Darwin's finches are an adaptive radiation, though a modest one. That's what I said already. As for your abstract, what evidence is there that this "small Komodo Dragon" is descended from a larger lizard? Or was that merely a descriptive term without evolutionary implications? Think about it.

Stanton · 13 July 2009

John Harshman said: Yes, Darwin's finches are an adaptive radiation, though a modest one.
"Modest" is too modest a word to describe any adaptive radiation that produces a vampiric species.
As for your abstract, what evidence is there that this "small Komodo Dragon" is descended from a larger lizard? Or was that merely a descriptive term without evolutionary implications? Think about it.
One recent report, mentioned in the wikipedia article about Megalania, suggests that the Komodo dragon is related to the Lace Monitor and Megalania.

John Kwok · 13 July 2009

John,

Don't tell me that you deny that there have been substantial adaptive radiations within cichlid fish in the great East African Rift lakes.

Thanks for admitting that the Galapagos Finches represent an adaptive radiation, since I believe that's been one of the scientific discoveries confirmed by eminent evolutionary ecologists Peter and B. Rosemary Grant as noted here:

http://chiron.valdosta.edu/jbpascar/Courses/Biol1010/ExtraCreditActivities/American%20Scientist%20Online%20-%20Adaptive%20Radiation%20of%20Darwin's%20Finches.htm

Anyway, I am merely reciting from memory the observation that the Flores Island hominid apparently hunted smaller versions of the extant Komodo Dragon, which, apparently, the authors of the scientific paper in question concluded were dwarf versions of the Komodo Dragon.

John Harshman · 14 July 2009

John Kwok said: John, Don't tell me that you deny that there have been substantial adaptive radiations within cichlid fish in the great East African Rift lakes. Thanks for admitting that the Galapagos Finches represent an adaptive radiation, since I believe that's been one of the scientific discoveries confirmed by eminent evolutionary ecologists Peter and B. Rosemary Grant as noted here: [here removed]
Please, please, please stop. But if you won't stop, please, please, please read for comphrehension. I won't tell you that rift lake cichlids aren't adaptive radiations, because they are. And I can't imagine why you suppose I have ever disagreed about this, or that it's at all relevant to anything we have been discussing. Nor have I ever disagreed that Darwin's finches are an adaptive radiation, so there's nothing to admit. However, you had previously claimed that there were several other adaptive radiations among Galapagos birds, and that I did dispute. Nor did the Grants confirm that Darwin's finches are an adaptive radiation. That had been known, and uncontroversial, for more than a hundred years before their work began. Their work concerns mainly the nature of speciation: sympatry vs. allopatry, the contribution of natural selection, etc.
Stanton said: “Modest” is too modest a word to describe any adaptive radiation that produces a vampiric species.
"Vampiric" is much too strong a word for a minor, opportunistic behavioral adaptation, which may or may not have a genetic component. Call me when they get fangs. If you compare the breadth of adaptation to various other island passerine radiations, e.g. Hawaiian honecreepers or Malagasy vangas, it's not much.

BNorman · 14 July 2009

I very much appreciate, and echo many of the comments made above. I work as a publicist for a publishing company which produces scientific books and journals (and shall remain nameless). So it is people like me who write the press releases which journalists use to write such shoddy science.

Needless to say, the team I work with make every effort to ensure each of our releases are checked and approved by the author before being sent. Thus ensuring his/her work is not taken out of context or sensationalised. (This makes sure we’re not behind stories such as this shrinking sheep nonsense.)

Often the blame for such shoddy reporting is placed at our door – after all the majority of people who work in the media or PR sectors have a humanities or social science background (in my case history) and so we are denounced for lacking the qualifications to write about science. To an extent this may be true – but we only exist because many scientists (clearly not all) lack the ability to write about their scientific discovery/study in a way in which the public finds engaging or understands. Our job is to act as translators if you will – taking a study or a discovery, which may be groundbreaking in its field – and making the public both aware and interested in it. I think this is a worthy cause, and a great aid to science – but shoddy reporting like the aforementioned story besmirches our name! There is a strong argument to suggest that media training should be encouraged more in the scientific community to that scientists can represent themselves and the opportunity to sensationalise or spin is removed - however such training would make me redundant so I won't push that too far!

If there is to be blame – and there should be - it should be laid at the door of some journalists who take press releases, run with them, don’t research them and simply hype up a particular angle. A recent paper we sent out was covered by 142 media outlets – and only 7 of those actually requested to read the paper. The majority just took the words from the press release and put their name at the top. That’s not journalism – that’s plagiarism.

So three cheers to Ben Goldache’s ongoing crusade against shoddy reporting and three cheers to all of you who hold all of us in the media/pr world to account!

RBH · 14 July 2009

BNorman said: I very much appreciate, and echo many of the comments made above. I work as a publicist for a publishing company which produces scientific books and journals (and shall remain nameless). So it is people like me who write the press releases which journalists use to write such shoddy science. Needless to say, the team I work with make every effort to ensure each of our releases are checked and approved by the author before being sent. Thus ensuring his/her work is not taken out of context or sensationalised. (This makes sure we’re not behind stories such as this shrinking sheep nonsense.)
[SNIPPAGE] Thanks for that. I'll voice a suspicion here: I suspect that no little part of the strange reporting of this particular study was due to the original investigators. At base, what they showed was that if strong selection for a phenotypic trait, in this case higher sheep weight, is eased (as it was by the warmer winters and longer breeding season, which permitted lower birth weights of lambs to younger smaller mothers and slower growth rates in lambs), then the distribution of that trait in the adult population shifts downward. It's noteworthy that while the authors provide the standard error of estimate for the various means, they provide no data on the actual population dispersion, say 10th and 90th percentiles or ranges. So we don't know if the decreasing mean weight reflects an extension of the left tail of the distribution or a shift in the position of the whole distribution with both tails shifting leftward. It's also clear in the paper that the authors regard "evolution" as meaning just adaptation under selection. For example, they write
Population density and maternal body weight (which is determined by the birth weight of the mother and her growth since birth) explain significant amounts of variation suggesting that any response to selection contributes substantially less than 4% to the phenotypic dynamics, suggesting that the recent dynamics of body weight have not been strongly influenced by selection and adaptive evolution. (p. 2)
That's pretty clearly an indication that for them "evolution" means just adaptation under selection, which is a very constipated conception of the process. As I remarked earlier, somewhere Larry Moran is gnashing his teeth.

Bill Bigge · 14 July 2009

Toidel Mahoney said:
Bill Bigge said: big is best - that explains why the only insects around these days are the size of cows and why bacteria don't exist! I agree though,lots of little bits of bad science reporting keep appearing and it can be frustrating, I sometimes try adding comments but not every story has a comments section.
Yes, it seems evolutionists like PZ Myers don't like that evolution might make predictions that could be falsified. Since evolutionism teaches that complexity increases through domination, and hence the bigger and stronger who are most successful at killing ultimately win. If the opposite sometimes happens, why can't we consider something other than Darwinism is going on? That question answers itself.
Umm, I guess you had your irony detector switched off when you were reading my comment. 'Evolutionism' says no such thing. Keep your fantasy world inside your head, we are not interested in it here.

RBH · 14 July 2009

I should add that as I see it, the value of the study lies in its explication of the way the shift in the population occurred when selection was relaxed, involving several factors, e.g., smaller lambs born to younger mothers and slower lamb growth rates. The blather about "evolution" being a minor contributor is simply confusing and even misleading.

John Harshman · 14 July 2009

RBH said: It's also clear in the paper that the authors regard "evolution" as meaning just adaptation under selection. For example, they write
Population density and maternal body weight (which is determined by the birth weight of the mother and her growth since birth) explain significant amounts of variation suggesting that any response to selection contributes substantially less than 4% to the phenotypic dynamics, suggesting that the recent dynamics of body weight have not been strongly influenced by selection and adaptive evolution. (p. 2)
That's pretty clearly an indication that for them "evolution" means just adaptation under selection, which is a very constipated conception of the process
I don't see it, at least not from that paragraph. If they had left out the word "adaptive" in the last sentence, you would have a case. But they didn't. Now I would consider "adaptive evolution" to be nearly synonymous with "selection", in that selection (possibly complicated by multiple levels) is the only mechanism of adaptive evolution we know of, though it's obviously not the only mechanism of evolution.

RBH · 14 July 2009

John Harshman said: I don't see it, at least not from that paragraph. If they had left out the word "adaptive" in the last sentence, you would have a case. But they didn't. Now I would consider "adaptive evolution" to be nearly synonymous with "selection", in that selection (possibly complicated by multiple levels) is the only mechanism of adaptive evolution we know of, though it's obviously not the only mechanism of evolution.
Well, let's look at the Abstract:
Environmental change, including climate change, can cause rapid phenotypic change via both ecological and evolutionary processes. Because ecological and evolutionary dynamics are intimately linked, a major challenge is to identify their relative roles. We exactly decompose change in mean body weight in a free-living population of Soay sheep into all processes that contribute to change. Ecological processes contribute most, with selection—the underpinning of adaptive evolution—explaining little of the observed phenotypic trend. Our results enable us to explain why selection is not realized even though weight is heritable and why environmental change has caused a decline in Soay sheep body size.
They're distinguishing between "evolutionary processes" and "ecological processes," where the latter (in their model) refer to the change in environmental conditions that relaxed selection. "Selection is not realized" because the selective environment changed, reducing the pressure favoring larger animals. I do not see how that can be contrasted with "evolution," as they do in the first and second sentences of the Abstract. Granted, a few sentences later they specify "adaptive evolution," but the die is cast in the first two sentences. Again, their paper identifies the mechanisms that yielded the shift in mean population weight under relaxed selection, but I don't see how that contrasts with "evolutionary processes" except in the narrow construal of strong directional selection.

John Harshman · 14 July 2009

You may indeed be right, and I have so far been too lazy to look up the actual paper. I merely say that the paragraph you quoted does nothing to support the claim you made for it. Saying anything more would require me to read the paper, which I really want to avoid.

RBH · 15 July 2009

John Harshman said: You may indeed be right, and I have so far been too lazy to look up the actual paper. I merely say that the paragraph you quoted does nothing to support the claim you made for it. Saying anything more would require me to read the paper, which I really want to avoid.
And it may be a bit of a stretch on my part. But something common in either the paper or the original public presentation of it in London led to some consistent mis-representations in the press.

Paul Flocken · 15 July 2009

Rilke's Granddaughter said: Henry, you just don't have anything intelligent or entertaining to say, do you. Poor boy.
Henry J said: Well, sheep are covered with wool, and wool can shrink when washed, so maybe these sheep were bathing more often than is typical for sheep?
Henry, If it is worth anything to you, I thought it was funny. :)

Kenneth Chang · 23 July 2009

RBH said: Scanning through some of the articles linked from that site, I see Kenneth Chang of the NYTimes got taken in too:
That was somewhat surprising, as larger sheep have better odds of surviving and evolution tends to favor those that are stronger.
I just noticed that I was mentioned in this thread, so I'll take this opportunity to say: I knew what I was writing, and what I wrote was factually correct. These are sheep on a remote, cold North Atlantic island, and the smaller, slower-growing sheep tend to starve to death during their first winter in this harsh climate. Hence, "larger sheep have better odds at surviving." I worded it to try to make clear that evolution does not necessarily select for larger body size, but that evolution selects for traits that help a species survive, which in this case happened to be larger body size. For what it's worth, what was written reflects what the researchers said. Here's the news release from Science. (emphases added)
Changing winter conditions are causing Scotland's wild Soay sheep to get smaller despite the evolutionary benefits of having a large body, researchers report in a study that shows how climate change can trump natural selection. The results highlight how wide-ranging the effects of global climate change can be, adding further complexity to the changes we might expect to see in natural populations in future. The study will be published online by Science, at the Science Express website, on 2 July, 2009. Science is published by AAAS, the nonprofit, international science society. "It's only in the last few years that we've realized that evolution can influence species' physical traits as quickly as ecological changes can. This study addresses one of the major goals of population biology, namely to untangle the ways in which evolutionary and environmental changes influence a species' traits," said Andrew Sugden, Deputy and International Managing Editor at Science. "Sheep are getting smaller. Well, at least the wild Soay sheep living on a remote Scottish island are. But according to classic evolutionary theory, they should have been getting bigger, because larger sheep tend to be more likely to survive and reproduce than smaller ones, and offspring tend to resemble their parents," said study author Tim Coulson of Imperial College London.

RBH · 23 July 2009

Thanks for your comment, Kenneth. A couple of remarks. The last paragraph of the news release reads
“Sheep are getting smaller. Well, at least the wild Soay sheep living on a remote Scottish island are. But according to classic evolutionary theory, they should have been getting bigger, because larger sheep tend to be more likely to survive and reproduce than smaller ones, and offspring tend to resemble their parents,” said study author Tim Coulson of Imperial College London. (Your emphasis)
But the paper at issue provides an analysis showing that selection for size was relaxed by the changing environmental conditions and the evolutionary benefit of larger size was reduced, and therefore the mean size decreased because younger ewes could reproduce, producing smaller, slower-growing offspring that could themselves survive and reproduce. While your article may accurately reflect the news release, the news release does not accurately represent the paper. What they showed in no way contradicts "classical evolutionary theory," and their characterization of it misled not only you, but virtually every other science writer who reported on it. As I said above
Now, it’s not clear to me that any of this is not evolution. It seems clear that the changed length and warmth of the summers eased selective pressure for larger sheep, permitting smaller sheep to successfully reproduce. Since the authors don’t report genetic information, we don’t know the nature of any change in the genetic structure of the population, but since body weight is highly heritable it seems obvious that it would have occurred. So the message in the research for me is that what happens to phenotypes is the net of the several variables operating in different ways to generate changes. It doesn’t make sense to me to say that “evolution” refers only to adaptive phenotypic changes that are due to directional selection for pure survival due to one environmental variable. In particular, I’m not at all clear on what it means to classify a “reproduction-related term” (OMD) as not evolutionary, as the authors did. It’s that reproduction-related variable that generated the size decrease.
Changing reproductive success as a function of changing environmental conditions is "classical evolutionary theory." Very early in the thread stevaroni wrote
Seems weird that two different news organizations would make the same mistake. I wonder if there was a press release from Science that tried too hard to sell a sexy evolution angle and ended up muddying the waters.
Just goes to show you can't trust news releases for an accurate rendering of the research, even when the release quotes one of the authors.

Morgan-LynnGriggs Lamberth · 24 July 2009

All this lends support to George Gaylord Simpson's statement in " The Life of the Past" that there is no orthogenesis- no predetermined path, just what natural selection can without teleological direction work. Whilst Cope's rule and others work, they do on the basis of what selection and randomness in the form of mutations can effect rather than from what some teleological force would produce.
Thanks to you others who confirm that these writers need to better fathom evolution. With PZ Myers, Amiel Rossow @ Talk Reason and Jerry Coyne in " Seeing and Believing, I argue that Kenneth Miller, foremost evolutionist and defender of evolution himself, sells evolution short!

Kenneth Chang · 26 July 2009

RBH said: But the paper at issue provides an analysis showing that selection for size was relaxed by the changing environmental conditions and the evolutionary benefit of larger size was reduced, and therefore the mean size decreased because younger ewes could reproduce, producing smaller, slower-growing offspring that could themselves survive and reproduce.
Which is exactly what I wrote:
But thanks to changing climate, the survival of the fittest has become a bit easier, enabling more of the less fit to survive. Fall now lasts later into the year and spring arrives earlier, and more of the smaller lambs, which once perished in winter, now survive to their first birthday.
I have two points: One is that, unlike some of the other articles, I did not say this violated classical evolutionary theory and I did not say that evolution (always) favors the bigger, which were the primary complaints of "crummy science reporting." And thus I am defending myself that I was not "taken in." Like people commenting in this thread, I was confused enough by the news release that I interviewed two of the researchers trying to understand what they were claiming. Within the constraints of a 250-word mini-story, I think I got the story essentially correct. The second is that this thread could just as well have been titled "More crummy explanation of science by scientists" or "More crummy journal hype." It's easy to say, "It’s truly crummy reporting, distorted and sensationalized. But there are also the casual errors, little misrepresentations that slip into stories almost unconsciously." In this case, the reporting of the stories pretty accurately reflected what Science itself put out in the news release and what the researchers said during a news conference. (If anyone would like to hear that, email me at kchang@nytimes.com.) It's a fair criticism if you feel that the reporters should have done a better job of putting a fuller context around what the researchers said. But a good portion of how science is portrayed in the media is a result of how scientists portray their science to the media. (Glass houses and all that.)

RBH · 26 July 2009

Kenneth Chang said: [SNIP] I have two points: One is that, unlike some of the other articles, I did not say this violated classical evolutionary theory and I did not say that evolution (always) favors the bigger, which were the primary complaints of "crummy science reporting." And thus I am defending myself that I was not "taken in." Like people commenting in this thread, I was confused enough by the news release that I interviewed two of the researchers trying to understand what they were claiming. Within the constraints of a 250-word mini-story, I think I got the story essentially correct.
Re-reading your story, I'm in agreement: You appear to have gotten it essentially correct.
The second is that this thread could just as well have been titled "More crummy explanation of science by scientists" or "More crummy journal hype." It's easy to say, "It’s truly crummy reporting, distorted and sensationalized. But there are also the casual errors, little misrepresentations that slip into stories almost unconsciously." In this case, the reporting of the stories pretty accurately reflected what Science itself put out in the news release and what the researchers said during a news conference. (If anyone would like to hear that, email me at kchang@nytimes.com.)
And again, as this has developed I do not disagree. In fact, as I dig deeper into it, this is reminiscent of the "Ida" hype. And I understand you're constrained by space. So I'm happy to concede that (a) within the space constraints you did a fair job of reporting, and (b) your suggested re-titling of the post is a legitimate suggestion.
It's a fair criticism if you feel that the reporters should have done a better job of putting a fuller context around what the researchers said. But a good portion of how science is portrayed in the media is a result of how scientists portray their science to the media. (Glass houses and all that.)
Agreed.

RBH · 26 July 2009

Kenneth kindly sent me the audio of the London press conference announcing this paper. First, from the introduction of the research by Andrew Sugden, International Managing Editor of Science International:
In changing environments, ecological and evolutionary dynamics are quite intimately intertwined. Traditionally, we think of ecological change as happening on shorter time scales than evolutionary changes, although some recent studies are also showing that evolutionary change can be very rapid. So if we observe the changes that can occur in some aspects of an animal's morphology, say such as body size, changes occurring over a number of generations, can we tell whether these changes are ecologically based or evolutionary or some combination of the two? In the research that's published on ScienceExpress today, Arpat Ozgul and his colleagues have used data from a wild population of Soay sheep, where mean body size has fluctuated substantially over the past 25 years but has on average got smaller, and this is despite the fact that there's positive selection for larger body size. Using a new body of theory they've shown the ecological response to environmental variation is the major driver of these dynamcis and evolutionary change has contributed relatively little. And they further explain why positive selection for body weight doesn't result in an increase in size, and the sheep have got smaller because climate change has modified the way that competition for forage, which depends on the density of the population, influences the growth rates of the lambs.
Then Tim Coulson, one of the authors, gave an overview of the paper, saying
So what motivated us to do this research is biologists have been realizing that many species, or several species of animal, can change their morphology relatively rapidly over a few generations. And [inaudible] theory that we've used are unable to explain this. There is also being (sic) several predictions where people would have expected animals to have increased in size but have remained the same size.
The right question got asked at 8:18:
Question: I'm not entirely understanding why you're calling the change ecological instead of evolutionary. It seems like what you're talking about is something like a lifting of evolutionary pressure, is that right? Coulson's answer: It's really been a change in the way that selection and ecology have played. So selection is the, as I say, the underpinning of evolution, and what this means is that individuals of a certain size are more or less likely to survive or reproduce than individuals of other size. And what we've found is that as the winters have got shorter the strength of selection has been reduced a bit. But the growth aspects, how fast or how slowly the surviving individuals are growing, is, you know, slightly different from selection. And that's what we're referring to as the ecological process. However these two processes do interplay with one another, so the growth rates of individuals made throughout their life again to influence [I think that's what was said there, though it sounds clumsy] whether they are likely to survive or to reproduce in a good or bad year. So that's why we're kind of really stressing is an interplay between the ecological and the evolutionary processes. And what we've been able to do is develop a method that allows us to tease apart the contribution of selection, the evolutionary bit, and the contribution of growth, the ecological bit, and to work out how they actually feed together and interact and how climate change has been influencing both of those processes.
"Slightly different from selection"? Yeah? How? Is the only difference between "ecological" dynamics and "evolutionary" dynamics how fast change occurs? That's the only difference offered in the press conference by both Sugden and Coulson. If that's the case, then we should be describing the Grants' research on changes in beak size of Galapagos finches as a function of annual rainfall "ecological" change rather than an example of natural selection in action. I note that the Ozgul, et al. sheep paper doesn't reference the Grants' work, though they saw equally rapid morphological change in response to changes in the physical selective environment. What actually happened on that damned island? The mean size of the sheep population, though highly variable, decreased over time. Why did it decrease? The environment changed, thereby easing the selective pressure against smaller sheep. What specific mechanisms/processes produced the smaller sheep as the selective pressure against smaller sheep eased? More smaller ewes survived and successfully reproduced, producing smaller lambs with slower growth rates, the latter (at least in part) due to increased population density and consequent competition for forage. Once again, as far as I can see Ozgul, et al. have described an evolutionary process and have identified the particular variables that account for the specific evolutionary change seen on that island. And that's it. I happily grant that separating out the contributions of several variables (e.g., ewe breeding size vs. lamb growth rate) was ingeniously accomplished, but I can't see anything that's gained by calling the latter variables "ecological" and claiming some useful difference between "evolutionary" and "ecological" change in the morphology of the population. No wonder those poor suffering writers were confused. Can anyone help me here? What, if anything, am I missing?

Henry J · 26 July 2009

So does "ecological change" mean that the sheep are smaller because they got less (or lower quality) food while growing up?

Henry