The idea of ecological communities has changed tremendously over the past forty years. The classical view stated that there were so many different species because evolution packed them tightly into the available niches. The modern view emphasizes the idea of trophic cascades, or top-down control in food chains. This emphasized the importance of predation in ecology, although it downplayed the significance of food webs, which showed the interrelated nature of ecosystems better than simple food chains.
Cotgreave, Peter and Irwin Forseth. Introductory Ecology, chapter 11
April 6, 2009
Professor Stephen Stearns: Okay, today we're going to start talking about ecological communities; and this builds on our prior discussion of competition, predation and disease.
So I'm first going to give you the classic view; by classic, I mean the view of community ecology from about forty years ago. Then I'm going to talk about something that has become a very dominant and influential point of view more recently, which is top-down control and trophic cascades. Then I'm going to emphasize the importance of history. So communities are not some kind of isolated microcosm, they develop in the real world and they have a history, and that's important. And they also exist in space.
So next time, actually, we're going to talk a bit about geography, island biogeography and metapopulations. This is making kind of a jump ahead into that. Dispersal really does make a big difference to the structure of communities; and I'll mention that briefly in this lecture.
And the take-home message really is that communities, which consist of basically all of the species that you find in a local habitat, and how they interact with each other, have been shown to be shaped by all of these things: competition, predation, parasites and pathogens; so in a real system they all interact. And it is one of the aims of community ecologists, out of this complexity to pull a few take-home messages that they can apply broadly. So we'll be discussing some of those today.
So the classical view on competition-driven species packing, on the planet, was developed by Evelyn Hutchinson. And the key reference is from just fifty years ago, "Homage to Santa Rosalia." Santa Rosalia was a saint who had a church named in her honor in Sicily, and Hutchinson visited this church on vacation, and in a pool in the church he observed two aquatic bugs. And one of them was large and one of them was small, but otherwise they looked the same and they ate the same things.
And so that led him--it's an amazing generalization, isn't it? You're kind of wandering through the churchyard and you see a pool and it's got two species of aquatic bugs in it--and immediately Hutchinson's mind jumps out to the whole planet, and wonders why are there so many different kinds of plants and animals? And his answer is because evolution has packed them tightly into the available niches.
And if it has done so, then there should be a limit to the similarity of competing species; which is what he had seen. He'd seen these two species living in this pool, and they're not the same size, they're different. And he's wondering what is the evolutionary process which has caused them to have different sizes? And he thinks it's competition.
And so he measured them, and then he went back and did a literature review and came up with the claim that the body size of the smaller one seems to stabilize at about 75%, as a length measurement of the body size of the larger one. And he goes into this. This is from the paper. And you can see here that the kinds of things he's looking at are mustelids; so these are weasels, and this is a mouse, and then these are two nuthatches in Iran, and then these are Galapagos finches.
And from that he gets a measurement when allopatric and a measurement when sympatric; and basically the summary of all of this is that species differ more when they're living together than when they're living apart. It's as though when they are living together, they're being pushed by something to be different. Okay? And that was taken as an estimate of limiting similarity under competition. So that's a classical view. And you've seen previously some of the data; you've read about some of the data that indicate that competition is important. A lot of that is field manipulation experiments. So that's one idea: The world is the way it is because it's shaped by competition.
The second idea is this idea of top-down control. By top-down, basically the top is the top of the food chain, and down is the bottom of the food chain. So this is about predators eating things in a food chain. And in particular it can interact with competition.
So one of the early ideas was that if a top predator prefers to eat prey that are competitively superior, then predation enables the competitively inferior species to survive.
So what would happen is if you pulled out the predator, so there was no top level in the food chain, and you just had say the herbivores competing with each other--the things that would be eating algae or plants competing with each other--then you would find that the dominant competitor would exclude the inferior competitor, which would go extinct, and you would have a simpler community.
So in this sense predation is maintaining biodiversity in the community, and it's making possible for the inferior competitors to do well because they're better at escaping the predator.
So the key papers on this are Brooks and Dodson; a 1965 paper that was done here, by two people in this department. Stan Dodson was a sophomore when he helped John Brooks do this paper. And they did it on data on daphnia and other zooplankton, living in reservoirs and lakes in Connecticut, in which there were alewives; fish that came in, or were landlocked, that would be eating some of the--you'll see the picture in a minute--that were eating the zooplankton.
And the other classical paper is from Bob Payne. That came out thirty-three years ago. And what Bob did was work, like Joe Connell, in the rocky intertidal. Bob worked off the--he worked on Tatoosh Island, on the northwest coast of the United States; it's on the Olympic Peninsula.
And there he could remove starfish from the intertidal, and he did so sort of in blocks in the intertidal. And he could go out frequently enough so that he could really keep the starfish off, and then he looked at how the community reorganized itself when he removed this top predator.
Now energy is flowing up through trophic levels in the food chain. And when we talk about top-down control, basically we are talking about how the secondary carnivores, or the primary carnivores--well the secondary carnivores are eating the herbivores, and the herbivores are eating--excuse me, the primary carnivores are eating the herbivores; the herbivores are eating the plants that are trapping solar energy. And when we talk about top-down control, we're basically talking about either the secondary or the primary carnivores, usually controlling what's going on below them in the food chain, going up this way, and the omnivores and the detritivores are usually off in another loop.
So the food chain idea, that you go up through a simple linear chain, is actually quite a simplification. If you look at a food web, then things get a little bit more complicated. This is a food web that you might find on the north slope of Alaska, and it contains birds, marine animals, foxes, things living in the soil, decaying matter. And you can see that there are potential flows of matter and energy in quite a different number of different directions in this food web.
So when we simplify that picture and say there's top-down control, I just wanted to have that in the background. I don't have a food web for this particular pond here. This is Lindsley Pond and other such ponds and lakes in Connecticut. But in Lindsley Pond, what Brooks and Dodson showed is that if you have an alewife--I'm going to just go out of this for a second, go up here, and go back in--if you have an alewife in the lake, and you have an array of zooplankton, of different size, okay? This--that is--there we go.
This is the size distribution of things in the lake, when there are no alewives. You put an alewife into the lake and you shift the size distribution down to the smaller size categories. So the idea behind this is that when there are no alewives, the bigger guys are out-competing the little guys. You put the alewife in to eat the big guys, and it gives the little guys a chance.
And basically the ones that are taken out are Daphnia and this Mesocyclops; which is a copepod. And the ones then which are allowed to increase in frequency are little things like Ceriodaphnia; that's plankton, which is a rotifer; Bosmina, which is another cladoceran, a relative of Daphnia; and Tropocyclops, which is a small copepod. So simply by putting this fish into, or taking it out of a lake, you completely restructure the community.
And remember that big things take longer to mature, and small things mature more rapidly. So what's going on here is not just a shift in the size distribution of the species that live in the lake, it's a reorganization of the whole population dynamics and rate of energy transfer of things in the lake as well.
And as a Ph.D. in this department is currently showing, it actually alters, … the presence of the fish in the lake, causes an induced response in the life history of the zooplankton, and the induced response, which is basically that they shift a lot more energy into reproduction--not of growth, because they're going to get hammered by the predator--that causes them to reallocate their fat reserves in such a way that the ratio of elements in their bodies changes.
So these kinds of changes have very deep-reaching consequences that cascade through the entire community and affect not only energy flow, but the ratios of elements like carbon and phosphorus and nitrogen, that you find in different parts of the food chain.
In the rocky intertidal, the classical story from Bob Payne has to with Pisaster. Pisaster is a keystone predator. By the way, this is an image taken off of the web, which is in the public domain. Pisaster looks like this, but that's actually not Pisaster. Pisaster is a bit oranger. These are mussels, down here; and these are the barnacles, over here; and these are some predatory snails.
So what's going on in the rocky intertidal is that there's a starfish which is eating snails, and it is also eating limpets, like that; and it's eating mussels, like that; and it's eating barnacles, like that. And if you take out the starfish, which prefers to eat mussels, then the mussels exclude everybody else; you just get a forest of mussels that covers the intertidal. And you put the starfish back in, and it will eat up the mussels and clear space that allows the others to exist. So this gives you the general starting idea of a trophic cascade.
So those ideas really come from Brooks and Dodson's observations on a natural experiment with alewives, and Payne's manipulation experiments with starfish. And the effects of predation are seen, under this hypothesis, as propagating all the way through food chains or food webs.
There's a key summary paper here, if you're interested in this. A way to get an anchor in the literature, and then work forward and backward, through Web of Science, is Pace, Kohl, Carpenter and Kitchell. And they define a trophic cascade as an interaction that yields an inverse pattern in abundance or biomass across more than one link in a food web.
So, as an example, if you have a simple three-member food web, if you have abundant top predators, they are reducing the numbers of the middle level consumers, and they're increasing the biomass of the basal producers. So the enemy of my enemy is my friend, is the idea behind a trophic cascade.
We have an evocative quote about this. So this is Aldo Leopold, one of America's great conservationists, who graduated from the Yale School of Forestry about eighty years ago.
Student: A hundred.
Professor Stephen Stearns: A hundred? Is this 100-- is it 100 this year? I'm getting older, because I actually visited Aldo's house in Wisconsin at one point. So [laughs]. I'm trying to avoid the implications of my own age.
So he lived at a time when park naturalists rode around on horses, and he was riding through parks, in various parts of the United States, when he made the observations on which this quote is based. So:
"Since then I have lived to see state after state extirpate its wolves. I have watched the face of a many a new wolfless mountain and seen the south-facing slopes wrinkle with a maze of new deer trails. I have seen every edible bush and seedling browsed, first to anemic desuetude and then to death. I have seen every edible tree defoliated to the height of a saddle horn."
So that's the consequence of pulling the top predator out of a North American ecosystem. And these sorts of trophic cascades can be observed also when the logging industry goes through and clear cuts an area. If you have well-developed forest, deer are at fairly low numbers, and therefore mountain lions are also at fairly low numbers.
If you clear-cut the forest, there's kind of an inverse trophic cascade. The deer build up, because there's a lot of browse. The mountain lion population grows up, but then the trees grow up. And as the trees grow up, the deer go down, and then there are too many mountain lions.
So when I was in graduate school in British Columbia in Canada, there was--the forests surrounding Vancouver were in this state, at that time, and there were a bunch of hungry mountain lions that were out looking for food, and they would come into the suburbs of Vancouver and do things like eat dogs.
And one day a man went out and saw a mountain lion dragging his son off, and attacked the mountain lion with a boat oar that he happened to have, and drove it away. So these kinds of things are actually still going on. By the way, the boy survived, the mountain lion was shot. And it's a shame that humans and wildlife have to interact, at that scale; it's pretty hard on the wildlife and it's pretty hard on the humans.
Okay, here is a summary diagram of what one means by trophic cascade. So you have a carnivore or a herbivore and a plant. And this summary diagram actually introduces a new element. This is from one of Os Schmitz's papers. Os is a professor in the Forestry and Environment Science School.
And what Os is showing is that besides just changing the numbers of the population, at the intermediate level, the presence of a predator--the carnivore here--will cause a herbivore either to reduce its activity or to shift into a different habitat.
And he demonstrated this, by the way, with field manipulation experiments in which the top predator was a spider, and the herbivore was a grasshopper, and the plant was a grass or a forb.
So if you have a lot of spiders, they will cause the grasshoppers to reduce their activity, and that will give an indirect positive effect on the growth rate of the plants. And if you have some grasshoppers that prefer to eat one kind of grass, but if the spider is particularly good at foraging on that- in that kind of habitat, and you introduce a spider, then it will cause the grasshoppers to switch onto the less preferred plant, and this actually causes the community to reorganize.
So you shouldn't think that the effect of a trophic cascade is just on the numbers of the herbivores, or of whatever is at the intermediate level. It changes their behavior, it changes their habitat preference, and, as I indicated earlier, it changes their life histories--in lakes, they have induced life history responses--and it changes their ecological stoichiometry, it changes the ratios of elements in the freshwater system. So these trophic cascades really have a lot of effects on the rate and the nature of matter and energy flowing through communities.
Some of them are with great big photogenic BBC quality kinds of interactions. Killer whales like to eat sea otters, and sea otters like to eat sea urchins, and sea urchins like to eat kelp. And so one of the classic North Pacific trophic cascades has to do with the density of killer whales and whether or not they prefer to eat sea otters.
Now, it was observed in the 1980s, 1990s and early twenty-first century that killer whales were developing an increasing preference for eating sea otters; which, by the way, had previously been pretty much eliminated by the- from the system by Russian fur traders, and then had come back in. And sea otters are something that people love to observe. They're extremely cute. So they have some conservation cachet.
Do any of you have--given the context we're operating in, community ecology, anybody have an idea on why it is that killer whales decided to eat more sea otters? Was there something that they might previously have eaten, that wasn't there anymore?
Professor Stephen Stearns: Sardines. No, killer whales actually have a hard time handling sardines. They're a bit small for a killer whale.
Professor Stephen Stearns: Salmon, yes; Alaska salmon fishery crashed. Killer whales got hungry and started eating sea otters. That's exactly it. So these kinds of things are going on all the time, and predators can switch, and that's what caused this particular trophic cascade to become obvious. Okay?
The killer whale switched onto sea otters. The sea otters declined. Because the sea otters declined, the sea urchins went up, and because the sea urchins increased, the kelp went down. So this is from that paper by Carpenter and Kitchell and others.
And the take-home message from this--I certainly do not ever expect you to memorize something like that. It might be nice to go through and pick out one or two favorite examples, or it could be useful as a way to get an entry to the literature, if you wanted to write a paper on it.
But it's to show you that trophic cascades have been now identified in marine, freshwater and terrestrial systems. And there are some very interesting ones. For example, mantids affect insects which affect plants; lizards affect grasshoppers which affect plants; wolves affect moose which affect balsam fir; mosquitoes, mosquito larvae eat protozoa, which eat bacteria; etcetera. Okay? So there are a lot of them. And it turns out that they play an important organizing role in a lot of communities.
Okay so, so far basically we've seen that the initial view of how a community is structured--which is that everything in it is competing, and the number species in it are determined by competition--was then modified by experiments and observations that indicated that predators have an important influence in food webs, and that if predators are selectively eating superior competitors, then they will maintain biodiversity and they will cause a trophic cascade, in the enemy of my enemy is a friend for the lower things, kind of pattern. Any idea why predators might prefer the competitively superior prey?
Student: There's more of them.
Professor Stephen Stearns: There's more of them, and so over evolutionary time they may very well have been shaped by evolution to be really good at eating the things that were most common. Yes?
Student: I was going to say, in certain habits, they all might be larger than the successful subspecies.
Professor Stephen Stearns: That's right. And so if you are an economically clever predator, you might take the biggest packet of energy available. Right. And, of course, if being a good competitor actually makes you more vulnerable, for some reason--if you are a tub of lard that can't move very fast--then that might do it too. Okay, now given--that was just a little summary of where we are so far.
Now let's see what other kinds of things will affect community structure. Because all of this is very good in theory, but when you actually go out there in the real world and you look at a real world community, there are some important things going on that haven't yet been put into analysis, into the analysis, and one of them is history.
So during the last Ice Age, patterns were created in North America that have allowed people to reconstruct what's going on, and Margaret Davis did excellent work reconstructing the history of the reassembly of the forests of North America, after the ice went off. And Rosie Gillespie has been- is currently doing a similar kind of reconstruction of the assembly of communities in Hawaii. And in both cases there are probes into time. In the case of the deciduous forests, the probes are essentially pollen; oh, and I'll wrap up with Bob Ricklef's general perspective.
So what Margaret Davis did was she went over eastern North America; and you can see from the black dots where she took a lot of her samples. Okay? And these are places where there was a lake or a bog into which pollen was falling into fine layers that could be dated by year.
And if you have an undisturbed bog, you are able to go down and make a core in it and actually count each year, as you go back. So, of course, this kind of thing can also be confirmed by carbon 14 dating, and you can get a number of crosschecks on the accuracy of your reconstruction.
What you see is that at the time of the full glacial extent of the Wisconsin Ice Sheet, you'll notice that it probed down well into southern Indiana here, and actually it came--right under here it is out over Long Island. The shoreline had dropped, as water was withdrawn from the oceans into the icecap. We had tundra conditions in southern New York and in Pennsylvania, and you had a spruce pine forest that went all the way down into southern Georgia and north Texas, and then you had a kind of a refuge down here for deciduous trees.
And as you can see from this sketch over here on the side, if you wash the sediments out, pollen has the great advantage--if you're a palynologist--of preserving its shape very well. Pollen is really tough stuff. The plants have evolved a lot of ways of preserving their precious DNA, and that serves us well.
So this is--you're not looking at DNA here, you're looking at the various kinds of structures that pollen gets packaged in. The important thing is they're really well preserved in pollen, and you can tell the difference. Okay?
So when Margaret reconstructed these things, she discovered that the first things to come back north, after the ice melted off, were spruce and larch. And so this is basically how long ago it was that spruce got this far north: 14,000 years ago, 12,000 years ago; 10,000 years ago it was pushing up into Canada. Larch moving up at a fairly similar rate; up here getting a bit north of Quebec City, about 8000 years ago. So you can reconstruct that pattern for these conifers, and then you can also reconstruct it for hickory and chestnut. So hickory gets to Maine about 4000 years ago, and chestnut only gets there 1500 years ago.
Now, if you were sitting at any particular spot, like say Albany, and you lived for 10,000 years and you were just checking off well when did that one arrive, and when did that one arrive, and when did that one arrive, you would see them coming in at different times.
And, by the way, just as a note on the importance of disease in structuring ecosystems, at one point there was evidently an outbreak of a forest disease that pretty much cleared out an area that ran from Massachusetts to Minnesota. That happened about 10 or 12,000 years ago. So the forests just basically all died, and then they were re-established, and it probably was a disease or a forest insect outbreak that did it. That's also in the record.
So they come, in at any particular point, at different times. Now that's an important comment on the idea of ecological succession, because a modern community, if you were just to look at it in time, and you were to forget history, it would--you could take communities from across a broad range of North America and they would all have very similar species compositions; they would all have these different hardwood and conifer species in them. And it might be that you would postulate that--you'd have a working hypothesis--that these species had to be there in a particular mix in order to co-exist.
The historical record indicates that's not true at all. They had quite different sequences of assembly. So it wasn't as though one species prepared the way for the next. One happened along in one place, and another happened along in another place, and they all ended up making communities of a fairly similar nature.
In particular, the hardwoods came in, in a lot of different sequences. Some are going fast, some are going slow. So it looks like the tree species you see in an eastern hardwood forest can be assembled in an almost random order.
By the way, the rates of their dispersal are determined a lot by what kind of seeds they had. Something like a hickory or an oak is going to move north about at the rate at which squirrels bury seeds. So it's going to be about 50 or 100 meters per generation, max. Something like a maple, that has a nice little helicopter seed, can go north on the winds and can move hundreds of miles at a clip. So the take-home message from Margaret Davis's analysis of the eastern hardwood forest is species assemblages come together in a potentially fairly random sequence. It isn't as though one thing prepares the way for the next.
Now let's take a look at what's going on with spiders that are ecomorphs in Hawaii. And by the way, these are the happy-face spiders; so some of them have a little mark on them that makes them look like they're happy.
And what are marked here are a group of the green ecomorph, which is waikamoi; this one. And then there is a maroon ecomorph, which is this group up here. And we have a small brown one and a large brown one. So this one doesn't look too brown, but they're called the small brown and the large brown one.
So you can already see that there's something going on with these spiders that looks something like the lizards in the Caribbean, the anole lizards that have different ecomorphs in the same community, and that that pattern repeats itself on different islands. Okay?
Now the oldest island is Kauai; the second oldest is Oahu--of the major islands--and then Maui Nui--which is Maui-Molokai-Lanai--and which, by the way, was all one island during the Pleistocene--is a younger island. And then some of these things are on the Big Island, which is--the oldest rock on the Big Island is only about 400,000 years old. Kauai is about 5 million. Oahu is about 2.6 million. Oahu is 1 million something. Yeah, 2.6 million for one volcano on Oahu and 3.7 for the other. Kauai out at about 5 million, and Hawaii at about 0.4.
So there are a couple of messages, when you look at these patterns. And you should think of the spiders as having started from Kauai and then moved down the chain. And there are two things that can go on. You can either have speciation or you can have assembly.
The difference in those words means the following: If you have assembly, it means that a bunch of evolution went on and generated a bunch of species, which we have over here in a pool, and then they migrated; and then communities got put together, with pre-existing species. If you have speciation going on, it meant that an ancestor came in and that it speciated locally, on whatever island, and that a community got assembled out of the clade that developed, with the different species in it. So that's the distinction between assembly and speciation. Okay?
And with T. quasimodo, you can see a spider that evolved here, and then spread down the chain, so that this particular one evolved on Oahu, this brown one, and then became part of the assembly process in the other islands. It crops up all over the place. So this is where that particular species originated, and then it became part of these other communities by a process of assembly.
The second thing you can see is that the diversity of the communities, which is indicated when you have four of them present, appears to be highest--you only, you get that on Maui and you get it here, on Oahu--it appears to be highest in communities of intermediate age. On the Island of Hawaii, which is young, and on the Island of Kauai, which is older, the communities appear to be simpler.
And I must say that this is kind of a rescue process to get this pattern, because these--Hawaii is the extinction capital of the world, and so there are a lot of people who are working to document these patterns before it's no longer possible.
One of the important things that sort of confirms Hutchinson's view of species packing is that Rosie has never found two species that share the same ecomorph in the same place; so that locally these things sort out.
And the maximum number of spiders you're going to get locally is four. And they're going to be in these different ecomorphs. They're going to be using different parts of the habitat; they're going to be behaving differently; and they're going to be colored differently.
And if you go back and you ask yourself, "How are they related phylogenetically?" you will see that, for example, the maroon one can have a green ancestor or a dark brown ancestor, and the light brown one can have usually a dark brown ancestor, but over here it had a green ancestor.
So the phylogeny indicates that the ecomorphs are arrived at convergently, and the phylogeography indicates that both speciation and assemblage have put them together into communities; and the local pattern indicates that you're never going to get more than four.
Well, take now that vision of history, that was given to us by Margaret Davis and by Rosie Gillespie and by others, and let's ask ourselves, are there some general patterns? That's what Bob Ricklefs, the question Bob Ricklefs posed himself.
Here you have an eastern hardwood forest; to remind us of what we're looking at. This is in the Appalachians. The historical perspective says that there's--the historical and the spatial perspective put together suggest that there's no such thing as a local community. Every community on the planet has been influenced by large-scale processes in space and time.
The regional and the historical forces that are acting on communities are just as important as predation and competition; maybe even more so. So if you want to understand those processes, you have to analyze them at a scale that's big enough to see them. And therefore there is now a discipline which is called macroecology, that tries to study these things at the scale of the planet, if possible.
At that scale you can see that dispersal is integrating things across regions, and therefore it is causing the process of assembly to take place; dispersal is bringing the species in from other places. And we can trace that process, as you've seen with Rosie's phylogenetic tree, using molecular phylogenetics.
If you create a landscape in a computer, and simulate this process, there's an interesting pattern that pops up. This is from some of Michel Loreau's work. On the Y-axis we have species richness, and over here we have how much dispersal is going on; how much are things moving around the landscape? And we have a couple of different definitions. Okay? We have local diversity and regional diversity.
So local diversity is going to be how many species you can count in a local patch, and regional diversity would be how much can you count in a state? Local diversity say for a lake in Connecticut would be how many species are in Lindsley Pond? Regional diversity would be how many fish do we find in the State of Connecticut? Something like that, okay? So the local diversity first increases and then decreases, as dispersal increases. Okay, so that means that the uniqueness of each species assembly is going to start going up, as you increase dispersal from 0. And then as you start dispersing right across the whole landscape, you homogenize it. So the uniqueness of a local community goes down.
The regional diversity stays stable until local diversity hits a peak, and then the regional diversity starts to decline, along with the local diversity. So as you just increase dispersal more and more and more and more, you decrease the opportunities for unique species to exist in places where they're favored by isolation, and you get maximum biodiversity at an intermediate level of dispersal.
Well, if we now look at a real world pattern, rather than at a simulation model, and we ask, do we really see what we might expect to see in something like the Amazon Rainforest? Then what we see is that a tropical forest really isn't completely assembled by random dispersal, from some kind of regional pool of species. Some species are much more widespread than expected.
So this is the reference on that. And basically what you see, if you look across distance, and look at the fraction of the species that are shared between any two points--in Panama, which is a fairly small country, on this scale, you can see that as you go out, the fraction of species that are shared, say if you took it two points that are fifty or sixty kilometers apart, they're only sharing about fifteen percent of their species.
But if you go into Ecuador and Peru, and you look at the fraction that are shared, across the huge expanse of trees that are available in the Amazon Basin, you find that there are some trees which are shared at great distances. They've spread across the forest and they are right across much of the continent. So regional processes really look quite different when you take the map, like a rubber, and just stretch it out, so that things can live in a larger area.
And by the way, there's quite a bit of history going on here as well. In the Amazon Basin, during the Pleistocene, much of the basin was actually savanna--up until about 12,000 years ago it was fairly dry in South America--and there were refugia. There was a refuge up near the Andes in Peru and Ecuador. There was a refuge in the Atlantic Forest, and then there was another one that was up in the highlands of Venezuela. And that huge forest, which now covers a third or a half of South America, has expanded out of those refugia in the last 10,000 years. And you can see that some of those trees were probably able to move fairly long distances, fairly rapidly, to get there and cover much of the rainforest.
By the way, the rainforest of the Amazon is not yet in stable age distribution since the last glaciation. You should not think of it as a stable ecosystem. The thing comes and goes, and it's in a process of transition all the time.
Okay, today I've tried to give you a bit of the intellectual history of community ecology, starting with an idea of communities in some sort of stable equilibrium, determined by competition. Then the idea of top-down control, through predation and dispersal; predation giving top-down control, and then dispersal and assembly in a regional framework. And then the perspective of time, that communities are assembled through time, and that this occurs at a geological timescale, with glaciers coming and going, and at a longer time scale with continents drifting. I think that really the communities are driven mostly by the glacial cycle, rather than by the continental drift cycle. And that means that when you look at something locally, you have to be able, in your mind, to integrate across these levels when you form your questions, because they suggest important alternative hypotheses that you might want to test. Okay, next time is Island Biogeography and Metapopulations.
[end of transcript]