Evolution Of Darwin’s Finches: Integrating Behavior, Ecology, And Genetics

[John] Good afternoon. Thank you all for coming out on this beautiful Monday. I’m John Fitzpatrick. I’m director of the Cornell Lab of Ornithology. It’s lovely to see you all out here as spring sets in. This is a really momentous day for the Lab of Ornithology. It’s our inaugural Paul C. Mundinger Distinguished Lectureship, and we are very pleased on this occasion not only to have a very distinguished lecturer, but also the entire Mundinger family here. 

And you’ll hear from Tom in a minute, but I wanted to acknowledge right off the bat, right off the bat, the generosity of the Mundinger family for remembering the late Paul Mundinger who was a very avid recorder of bird sounds, and a very distinguished scientist in his own right. And with this lectureship and I want to acknowledge, and ask if they’re willing to stand Mary Mundinger, as well as Paul, 

[Applause] 

that’s Paul Paul, Jr. And Anne and Elizabeth Mundinger. 

[Applause] 

Thank you all. The Mundingers are all Ithacans. They were here as kids, along with Mary and Paul when Paul was getting his PhD here. And they have, as you might hear from Tom in a second, some fond memories of the old Lab of Ornithology building. 

Paul taught at Queens College in New York City. He’s a distinguished bird recordist. His major part of his life’s work on tape has been donated by the Mundinger family to the Macaulay Library, we just took a look at it. And that filled half of a whole library shelf, all those wonderful tapes. 

Mary is a former dean of the Columbia School of Nursing. She’s very distinguished career in her own right. Winner of numerous awards, sitting on panels and boards having to do with public health, with health policy, and with academic health. So it’s a very great privilege to have the entire Mundinger family here for this inaugural Mundinger lecture. 

And it’s a pleasure to introduce Tom, who is going to say just a few words about the family and Paul, and Tom is in the Department of Medicine at the University of Washington, Seattle. Tom.

[Applause] 

[Tom] Thank you very much. As Dr. Fitzpatrick said we’re all from Ithaca. In the 1950s my dad and his wife Mary started their family here. All four of the children were born at what was then called Tompkins County Hospital. So Ithaca is a home to us, to all of us. And that’s part of the reason that we wanted to be remembered with a, my dad’s legacy to be remembered with a lectureship here. 

A little bit about this lecture series, and with this being the first in the annual lectureship, so this is annual. We hope to see you all and a couple of your friends back here with a couple extra seats next year and the years to come. But he felt a very strong connection to this city of Ithaca, the University of Cornell, and especially the Lab of Ornithology. 

He received his PhD from Cornell in the mid-‘60s and as Dr. Fitzpatrick said went down to the New York area and did his post-doctoral fellowship at Rockefeller University, and then went on to Queens College. 

But in his professional life my father was both a teaching professor as well as a researcher. And in his research he worked both in the lab and in the field, and we’re going to be hearing a lot of exciting work, field work, here this evening. And again his work was in animal behavior, his index of animal behavior was bird song. 

And my dad felt it was a great honor to have his professional, or anyone’s professional legacy, associated with a lectureship like this. So he would be very proud to have members of the Ithaca community, to have students both graduate and undergraduate, to have faculty, fellows, staff, to have a lectureship that includes all this in this educational endeavor. 

So on behalf of the family we’d like to thank you all for being here and for your attention, and we’d like to thank Cornell University and especially the Lab of Ornithology for being hosts to this lectureship that honors my dad’s professional legacy. So we’re in for a real treat this evening. And thank you very much, and enjoy the evening. 

[Applause] 

[John] Thank you, Tom. And now it’s my great pleasure to introduce Irby Lovette who’s the Nancy and Larry Fuller Professor of Evolutionary Biology, and the head of the Fuller Evolutionary Biology Lab at the Lab of Ornithology who’ll introduce our speaker. 

[Applause] 

[Irby] Thanks, Fitz. I think I have the easiest job of the evening because you all know why you’re here. I think this is an introduction for someone who basically needs no introduction to any audience that is is paying attention to the history of evolutionary biology in the last 40 or 50 years. And I could stand here and give you a very very long list of professional honors won, and papers published, and books written. If you’d like to see that you know go online, look at the Wikipedia entry, it’s it’s vast. 

What I’m gonna do instead is I think give you just a couple of anecdotes that I think might frame Rosemary’s talk on a slightly larger context. And and one thing I want to acknowledge is that this this very room that we’re in right now is kind of steeped in the discussion of Rosemary and Peter’s studies of finches. 

This is the room where for more than 20 years we’ve taught evolutionary biology to many many thousands of Cornell students who have gone on to illustrious careers of their own. And and so the the example of Daphne finches has been covered in this room, I think you know, it’s probably still echoing off the walls in here in various ways. 

I know many of you here are actually in that class right now, or maybe alumni of that class in recent years. And so this is a special treat. This is the first time we’ve actually been able to bring the source of that information into this context. 

But it’s. And I think pretty much every evolutionary biology class around the country does something like that. So we’re not special in that regard, but but I will say you know it doesn’t stop at the college teaching level. 

Just this past weekend we were touring around a group of 20 I think junior undergrads from 20 different colleges around the United States, and and they were seeing the posters about Rosemary’s talking and saying Rosemary Grant is coming to Cornell? 

And and I said yeah, and and and and I thought to ask them just kind of on the fly, you know, how many of you 20 students heard about natural selection in Darwin’s finches not in college but in high school? 19 hands went up. 

And so this this touchstone example of natural selection happening in the wild has really permeated I think all of our thinking at all levels about how natural selection actually functions. And so it’s not just professional audiences and and you know you know the ivory tower folks that have learned about evolution through these examples, it’s basically our entire culture across the United States and maybe even across the world. 

So that’s the context by which Rosemary is coming to talk to us, but her, the kind of iconic example that I think most of those students would have heard about in their high school biology classes was only the beginning of a really really long and very very impactful set of studies that Rosemary and her husband Peter who’s here today as well did over the course of more than 40 years. 

And so the example that you may think you know if you took high school biology or college-level biology a while ago has only gotten better over time. And I think Rosemary’s gonna bring us somewhat up to date on that tonight. 

And then I just want to close with a personal anecdote. I had the very very good fortune myself of seeing a job advertisement when I was 20 years old to serve as a field assistant for Rosemary and Peter on Daphne. And I answered that advertisement and and they chose me as their field assistant. So I’ve actually spent six months on Daphne chasing finches under the tutelage of Rosemary and Peter. 

And so I had kind of a you know a front row seat to the to the way in which they do their science, which I think is unusual. Many many people know about their science, but few have seen it in practice. And and I want to just point out that for me it really is is is kind of an almost magical combination of the greatest theoretical broad-based consideration of fundamental principles married with something that I got to see firsthand, a really really deep understanding of this, of the the study systems in which they work. 

And it’s not just the finches I remember, I remember being tutored in in in plants like Portulaca and Cacabus and Ipomoea, you know the seeds that are so important to those finches, and a whole bunch of other things. And so I want to say that Rosemary is not only an extremely exceptional and enormously influential scientist, but she’s also a consummate naturalist. And I think it’s that marriage of traits that has made her work so so so impactful for so many people. 

And with that I ask everyone to join me in a rousing Cornell welcome to Rosemary Grant. 

[Applause] 

[Rosemary] Well, thank you very much. Thank you very much for the introductions, and this is just a huge honor for me to be here today on behalf of both Peter and myself to be giving the inaugural lecture. 

We’ve had a really enjoyable morning so far, speaking to lots of students, post-doctoral fellows, and faculty members, and I just thank you very very much for the invitation to come and to talk to you today. 

So I chose this title with “Evolution of Darwin’s Finches: Integrating Behavior, Ecology and Genetics” 

[Slide text: Evolution of Darwin’s Finches: Integrating Behavior, Ecology and Genetics 

Paul’s innovative research on canaries, house finches, gold finches, and siskin integrated behavior, ecology and genetics] 

because Paul’s very innovative research in a number of species of birds really did integrate behavior, ecology, and genetics. And this is where Peter and I have followed in the same footsteps as Paul. This, by Paul integrating behavior, ecology, and genetics 

[Slide text: Showed that song learning involves an interaction between experience and genetically determined neural circuitry; Photo: Paul C. Mundinger] 

he was able to show, and this is I think one of his papers that I most enjoy reading, he was able to show that song learning involves an interaction between experience and genetically determined neural circuitry. And this is a really innovative piece of work that he and his colleagues did. 

[Slide text: Evolution of Darwin’s Finches General Environment → (Genotype Variation → Phenotype Variation → Fitness Variation) ← Social Environment Fitness Variation ⇢ Next generation ⇢ Genotype Variation] 

Now Peter and I, the way we think of it is I think the same way Paul does. We think that genotype variation is translated into phenotypic variation translated into fitness all under the influence of the general environment and the social environment. 

And this is the environment of other species in the same area, and also interaction with species within, individuals within the same species. Now all of us I’m sure in this room whether we are hiking in the alpine meadows 

[Photo: Alpine meadow] 

or we are 

[Photo: Amazon rain forest] 

traveling in the Amazon jungles or 

[Photo: Coral and tropical fishes in a reef] 

diving beneath the surface of the sea are just amazed at the diversity of life on this planet.

[Slide text: How is Biodiversity generated? How do we study this process of speciation?; Photo: Hillside] 

How is all this biodiversity generated? How and why do species multiply? How do we even go about studying this process of speciation? Well we know from a lot of people’s works, and I’m thinking of not only Dolph Schluter but many of other people’s that 

[Slide text: Much diversity arises from Adaptive Radiations, that is the rapid diversification of lineages from a common ancestor 

Hawaiian honeycreepers- Over 50 species have evolved in 6.4 MY, Fleisher and McIntosh 2001; Image: Drawing of Hawaiian honeycreepers showing various beak shapes] 

much diversity arises from adaptive radiations, and that is a rapid diversification of lineages or species from a common ancestor. One of the most spectacular of course is the Hawaiian honeycreepers with their enormous plumage differences, their bill shape differences. And over 50 species of these have evolved from a common ancestor in the last six to seven million years. 

But unfortunately many many of them have now gone extinct. Now Darwin himself suggested that actually young radiations, he didn’t put it quite that way but that’s really how we would translate it today, that young radiations might be the best place to observe this. 

And there are many young very, um radiations, there’s they cichlid fishes, the the char in Iceland, but we chose 

[Slide text: -Young radiation -Might be possible to observe and measure phenotypic changes in contemporary time?; Photos: Darwin’s finches] 

Darwin’s finches as a young radiation. And we know now that this radiation um occurred in the last um one to two million years, where there are now 15, possibly 16 species that have all been derived from a common ancestor. And we thought it was this radiation where it might be possible to observe and to measure phenotypic changes in contemporary time. 

And the reason was that it had several, several things that were helpful for this. 

[Slide text: -Islands differ from each other ecologically and impose different selection pressures -Pristine -Dynamic; Image: Map of Galapagos Islands] 

First of all they occur in the Galapagos archipelago. These islands differ from each other ecologically and they impose different selection pressures on the finches. So it might be possible to see how these different finches have adapted to this. 

The other thing is that these islands, many of them are close to being pristine, never having been inhabited by human people, by humans. And in such that so that if we were able to observe and measure any changes then it would be as the result of natural conditions and not human-induced conditions. 

And finally it’s very dynamic. They occur astride the equator and as such they are subject to the El Niño-Southern Oscillation phenomenon, which means that there are some years when there’s an enormous amount of rain falls in these islands, El Niño years, and there’s huge productivity of plants and also finches, or land birds, and then this is interspersed with years that sometimes can last as long as two and a half years where the no rain falls at all, it’s a complete drought, and many of the land animals die. 

And it’s, and this sort of situation where we’re act, actually to see changes in phenotypes. 

[Image: Map of Galapagos Island with arrows labeled 1 to San Cristobal, 2 from San Cristobal to Española, then Española to Floreana and Floreana to Santa Cruz, and 3 from Santa Cruz back to San Cristobal] 

Um Darwin himself thought that the speciation might occur, and this is purely a car, a cartoon, but it’s, so he supposed that finches probably arrived from the mainland, arrived on an island in the Galapagos, met completely different ecological conditions, changed through the process of natural selection and evolutionary responses to natural selection events, built up a population, moved to another island with yet different ecological conditions. 

Step one and step two was repeated over and over again until eventually they came together in secondary contact. 

Now if the birds have been changed quite a bit during this process they could be completely different species. If they had not changed very much they might interbreed, or as Darwin thought if they had only changed a small amount then there might be competition for food and there could be subsequent divergence. And in what we now call character displacement.

And I’ll be coming back to this later. 

[Slide text: Different species are at different stages of becoming genetically incompatible -Lineages diagnosably different -Lineages cannot interbreed, genetically incompatible] 

Now another way of looking at this is that the finches diverge, a population could diverge here, and one line could go off onto another Island with different ecological conditions, and they would diverge to a point where they would differ and the lineages would be diagnostically different in morphology, and these divergence would continue until they got to the point where they, even if they did come together they would be genetically incompatible.

Now the point is here, we know from Allan Wilson’s work, that this point from lineages being diagnosably different to the point when they really can’t interbreed for the reasons of genetic incompatibility can last as long as 32, on average, 32 million years in birds, in fish, and in reptiles. It’s an order of magnitude less in mammals. 

[Slide text: -Lineages diagnosably different, Pre-mating barriers to interbreeding -Lineages cannot interbreed, genetically incompatible] 

But Darwin’s finches only one to two million years old would lie around about this point where lineages are diagnosably different but long before the point when, um they’re kept apart by genetic incompatibility. 

So we’d expect at this point that they’re kept apart by what we call pre-mating barriers to interbreeding. Things like differences in song, and in other birds the differences in plumage. Not so in Darwin’s finches with, about the plumage, but it would be pre-mating barriers to interbreeding. 

[Image: Map of Galapagos Islands] 

Now Peter and I have done work on, we visited all the islands in the archipelago, but we’ve done in-depth studies on two islands. One of them is Genovesa in the north and the other one is Daphne here in the center of the archipelago. 

And it’s a story of Daphne that I’m going to tell you today. 

[Slide text: Isla Daphne Major: a microcosm of Evolution; Photos: Isla Daphne, and four finches] 

Now we can think of Daphne as a microcosm of evolution. When we first went there in 1973 there were two populations of finches. One species was Geospiza scandens, a cactus finch, which is a cactus specialist. It feeds on on pollen and nectar from Opuntia flowers, and also the seeds. 

And then there’s this other very interesting finch Geospiza fortis, which is a granivore, and a much more generalist granivore, with a much blunder beak than scandens. Now we were particularly interested in fortis because as David Lack had explained in his book that fortis is highly variable in bill dimensions and in body size. The sort of variation that we see in human height is just about the same amount of variation. And we thought well this might be a very good species to examine changes that might occur. 

[Slide text: Outline -How do new phenotypes arise in response to changes in the environment? -What is the source of genetic variation and how is it maintained? -Origin of a new lineage -Insights into 3 models of speciation; Photos: Finch and Isla Daphne] 

So the outline of my talk is how the new phenotypes arise in response to changes in the environment, what is the source of the genetic variation that allows this, and how is it maintained? 

And then I’m going to tell you about one of the most exciting parts and absolutely unexpected findings in our study, and that is the origin of a completely new lineage which we followed from its inception up through six generations. And then I’ll bring this all together in a couple of slides, has given us insight into three models of speciation. 

[Slide text: Geospiza fortis, medium ground finch; Photo: Banded medium ground finch] 

So this is this variable Geospiza fortis or the medium ground finch, and one of the first things we did is we banded a lot of birds. We watched how they fed 

[Slide text: Mean beak depths of fortis observed feeding on small seeds only (S), small and medium seeds (SM), or seeds of all sizes (SML). Lines represent 95 percent confidence intervals; Chart: Bar chart of Diet Categories versus Beak Depth in mm, showing finches with larger beaks ate seeds of all sizes while finches with smaller beaks ate smaller seeds; Photos: Small and large medium ground finches] 

and look, and found out that the small members of the population, this is an example of the small one and this is an example of the large one, same same species, same population on Daphne. 

The small ones fed exclusively on small soft seeds, whereas the largest members of the population were able to eat the large, hard fibrous seeds and they also took a few small and medium-sized seeds as well. And then the medium-sized birds ate the small and medium seeds. 

[Graph: Year versus Beak Depth in mm of G. fortis, Grant P.R. & B.R. Grant, 40 years of Evolution, 2014, Photo: Medium ground finch] 

Now over the 40 years there has been an enormous change in beak depth in this population. So it began in ’73 this is the average beak depth of the population at that time and these lines on either side are the 95% confidence limits on either side of the mean. And so we had this enormous change here when there was an increase in bill depth and then it went along for about a few, eight years or so, and then it dropped down again to where it was before then it continued low for the next 20 years, and then there was this precipitous drop here. 

Now I’m going to tell you about all these three points. The common denominator of all these changes has been a drought when large numbers of finches died. So during the drought 80 and sometimes as many as 90% of the finches died. 

[Photo: Booby on a rocky island] 

So we knew that if we looked at evolution by natural selection 

[Slide text: Evolution by Natural Selection 1. There must be a heritable variation in the trait of interest 2. As a result of some environmental perturbation there is differential survival 3. In the next generation the offspring of the survivors must resemble their parents in the trait of interest] 

there have to be three factors. There must be heritable variation in the trait of interest, our trait of interest were bill size and shape and body size. As a result of an environmental perturbation, in our case it’s going to be drought, there was different, must be differential survival, and very importantly in the next generation the offspring of the survivors must resemble their parents in the trait of interest. And they will do this of course if the traits were highly heritable. 

So again one of the first things we did early on in our study was we 

[Graphs: Mid-parent versus Mid-offspring Beak Length for G. scandens and Beak Depth for G. Fortis showing an association between parents and offspring of the traits] 

caught a lot of birds, banded them, measured them, and then also their offspring. Waited till their offspring had grown up to adult size and measured their offspring, then looked at the association between offspring and parents to get the heritability. 

And we found that the heritability was extremely high in all the traits. And I’ve just given you bill depth here for fortis and bill length here for scandens. And you can see the heritability is naught point seven two. Zero would be no heritability, one would be complete heritability. But this heritability is very high, and again, and if we were to do this with human height, human height comes out as a heritability about point seven. 

So we didn’t have to wait long for the first perturbation. 

[Photo: Isla Daphne with vegetation] 

This is just one part of the island, this is what it looks like in a normal wet year. 

[Photo: Isla Daphne without vegetation] 

And then this is what it looks like in a drought, when there’s not a blade of grass, not over, no vegetation at all. And the finches have to scrabble around in the rocks to find the remaining seeds. 

Now what happened, this was the drought in 1977, what happened was that the small seeds quickly went out 

[Graph: Date (July ’75 to January ’79) versus Beak depth in mm, showing an increase in beak depth, Boag and Grant. 1981, Science; Photos: Tribulus seeds and medium ground finch] 

leaving the big seeds. So birds began to die. Over 80 percent of the birds died. And as they, the young, the small ones went out first, until the end of the drought we were left with just a few birds and they were all the large birds in the population. Very few of the small birds had survived. 

And the reason was that the large, hard seeds of Tribulus were the only ones remaining at this time, which only the large finches could crack.

[Graph: Natural Selection- Sampling Times (July ’75 to January ’79) versus Beak depth in mm, showing an increase in beak depth] 

So this was a natural selection event, but natural selection alone is not evolution. Evolution occur, is a change across generations because of the heritability. So when the rains came back again the few large finches that had survived bred and produced offspring, which when they grew up were large like their parents. 

[Graph: Evolution- Sampling Times (July ’75 to January ’79) versus Beak depth in mm, showing an increase in beak depth] 

So this was an evolutionary response to a natural selection event for the reasons that it’s the large birds that could eat the large, hard Tribulus seeds. 

[Graph: Year (1970 to 2012) versus Beak Depth, mm of G. fortis with arrow pointing to 1978 point. Grant, P.R. & B.R. Grant, 40 years of Evolution, 2014; Photo: Medium ground finch] 

So the popul, the population went up to this point here, in bill depth, and it would look like this if I’d use body size as well. Now the next, and it went on for about ten years when large birds produce large offspring. And then along came a 

[Graph: Bar graph of Rainfall, mm per year from 1973 to 2012 showing large amounts of rainfall in 1981 and 1998] 

perturbation in another way. And this was an El Niño event that produced a huge amount of rain. It produced over a meter of rain on this small island. And this was, this was the largest, the most severe El Niño effect in 400 years. And we could tell this by coral core data. And it completely changed the island from a small seed produce, from a large, hard seed producer to a small seed producer. And I’ll just show you this in pictures. 

[Photo: Isla Daphne with some vegetation] 

So this is what the island looks like in a normal wet year. And these plants on the ground, [Photo: Tribulus plant in flower]

 mainly Tribulus plants which produce these 

[Photo: Tribulus seeds] 

large, hard seeds which made the difference between survival and non-survival in the previous drought that I told you about. But it went on raining for eight months. The birds actually bred for eight months. 

[Photo: Isla Daphne with abundant grasses and other vegetation] 

And these Tribulus seeds were smothered by grasses and other herbs. It rained some more 

[Photo: Isla Daphne with vines growing on the trees] 

and they were covered by vines. Vines grew up 

[Photo: Vines growing over a cactus] 

over cactus bushes, they grew up over trees and then the next year after it stopped raining 

[Photo: Mostly dead vegetation, but vines still over trees and cactus bushes] 

you could still see the trees and the cactus bushes draped with these vines that covered the islands. And our quadrat data showed 

[Graph: Bar graph of Wet Mass mg per m2 of small and large seeds by year from 1976 to 1991 showing an increase in small seeds and decrease in large seeds] 

that the island had gone from a large, hard seed producer in green to an abundant supply of small, soft seeds. So when the drought came two years later it was this time, it was the smaller birds who had the selective advantage. And it was them that survived 

[Graph: Year (1970 to 2012) versus Beak Depth, mm of G. fortis with arrow pointing to 1985 point. Grant, P.R. & B.R. Grant, 40 years of Evolution, 2014; Photo: Medium ground finch] 

and we plummeted down to back where we were in bill depth. But when we look at bill shape a very interesting 

[Graph: PC2 Beak Shape by year, 1973 to 2012] 

thing happened. There was, at this point, there was this incredible difference in bill shape where these birds had all got blunt beaks, these were much more pointed beaks. 

[Graph: Beak depth versus Beak length for G. fortis, showing shift in allometry] 

And if we look at this in another way, you can see when we look at bill length over bill depth, this is for the whole 40 years. We started down here, we went up to this point here in that 1977 drought when only the large birds survived, and came down here and went across her. And it’s this that we get this change of allometry. 

So why did the bird suddenly have sharp, pointed beaks? Well the answer is gene flow between scandens and fortis at that point. 

[Slide text: What causes rare hybridization between fortis and scandens? Why did some hybrids between fortis and scandens survive to backcross in 1986?; Graph: PC2 Beak Shape by year, 1973 to 2012] 

And this was a result of hybridization between the sharp billed scandens and blunt beaked fortis, and then backcrossing to one or other of the parental species. So there was a trickle of this. Only about one [inaudible] percent hybrids were formed, but nevertheless this was enough to give a trickle of genes going from one species to the other. 

So what was it that caused this rare hybridiz—hybridization between fortis and scandens? And why did some hybrids between fortis and scandens survive to backcross in 1986 and not before? Well to answer this we’ve got to find 

[Slide text: What is the pre-mating barrier to reproduction between species? Why did it leak?; Photos: G. fortis, G. scandens, and a nest] 

what is the pre-mating barrier to reproduction between the species? And why did it leak at this time? So in all Darwin’s finches there is no plumage difference. Males are black, females are brown. They produce these dome nests. They have similar courtship displays as far as we can see. 

But they differ very much in 

[Slide text: Song and morphology] 

song and morphology. And so we recognize them by song and morphology, but can the birds? 

[Slide text: Can individuals discriminate between their own and another species -by morphology? -by song?; Photos: Male finch displaying to a stuffed female finch, and a tripod in a grassy area near trees] 

We had to ask can individual birds discriminate between their own and another species purely on the basis of morphology and the absence of song? And can they do it by song in the absence of morphology? 

So to find out whether they did, could discriminate between their own and another species on morphology we took stuffed museum specimens, we put a female fortis on one side of a pole, female scandens on another, took it into the territory and asked can the male territory owner discriminate between a female of its own species and a female of another species even though it was just a stuffed museum specimen? 

And the answer was a resounding yes, they vigorously courted, as you could see here, a female of their own species and completely ignored the other even though it was only a stuffed museum specimen. And of course we did this many times, and we did it with a control. 

And then we asked can they tell the difference by song? So we recorded song, played back fortis song, fortis came into the loudspeaker, scandens went on feeding or doing whatever it was doing. And played back scandens song, scandens came in, fortis went on feeding or doing whatever it was doing. And we again did this many times and with controls. 

So just as we can distinguish between the species with song and morphology so can the birds.

[Slide text: Individual variation between songs but always on a species specific song theme; Images: Spectrogram of G. fortis and G. scandens song and photo of each species] 

Now the songs are very different. Fortis song’s a modulated song here, and then scandens song is a repeated note song. There is individual variation but always on a species specific song theme. Now we know from work by Bowman that song, because he was able to do work with captive birds, you can’t do that now, but he was able to do this in the 1950s. 

[Slide text: Song is learned from the father in association with his appearance during a short receptive period early in life between day 10 and day 40 (Bowman 1961)] 

And he was able to show that song is learnt from the father in association with the appearance during a very short sensitive period of time between day 10 and day 40 after hatching. Now this time corresponds with the last few days when the birds are in the nest, and when they’re out of the nest being fed by their parents. And all this time the male is singing. 

So it’s not too surprising that they learn almost there, what perfect renderings of their father’s song. 

[Slide text: -Once learned the song is retained for life -As adults, they pair according to their species song] 

And once learnt this song is retained for life. We have recorded song repeatedly through the lives of the birds and it has not changed. And as adults they pair according to their species’ song. So this constitutes a pre-mating barrier 

[Slide text: Pre-mating barrier to interbreeding -Song: learned culturally transmitted -Morphology (beak and body size): genetically transmitted] 

which based on song, which is learnt and culturally transmitted, and morphology and particularly beak and body size, which I’ve shown you is genetically transmitted. 

[Slide text: -How robust is this pre-mating barrier? -It is vulnerable to disruption if a young bird hears and learns the song of another species during its sensitive period] 

But we can ask how robust is this barrier? Because after all it is vulnerable to a disruption if a young bird hears and learns the song of another species during this short sensitive period of time. And this does happen, it happens rarely about one percent of the time.

And it happens for different reasons, but one reason is if the male dies and the female is left to rear the offspring, and then females don’t sing, and if the natal neighbor is another species, then the birds in both those nests will learn that natal neighbor’s song, and so they will grow up learning the other species’ song. 

Now we had a few birds produced like this actually from the very first time we went to the island, just one at the most two percent. But it happened every breeding season. So we had a few hybrids that we could follow between fortis and scandens. 

[Photo: Isla Daphne with little vegetation] 

And none of the hybrids in the first 10 years of the study survived the dry season to breed. We thought maybe this was because they didn’t have enough of the appropriate food, lots of birds were dying at this time anyway. And, but we also thought maybe we could see the beginning of genetic incompatibility. But actually the first was right, that the birds with the intermediate beak size, there was not enough food available for them. 

[Photo: Isla Daphne with lots of vegetation including grasses] 

Because after the El Niño of 1983 when there was this large production of small, soft seeds then the hybrids began to survive, and they backcrossed to either one or other of the parental species according to the song that they had learnt. 

[Slide text: G. fortis and scandens converged morphologically; Graphs: Beak length versus beak depth of G. fortis and G. scandens in 1975, 1987, 1991, and 2012] 

So fortis and scandens, again this is a plot of bill depth over bill length, but it shows you that for the first ten years of our study fortis and scandens were completely separated. And then when there was this trickle of gene flow from fortis into scandens and scandens into fortis, the populations started to converge on each other. And this convergence continued for the next thirty years. 

[Slide text: Were there specific alleles introduced from scandens into fortis that could have contributed to fortis more pointed beaks?; Graph: PC2 Beak Shape by year, 1973 to 2012] 

So this was what was producing the, the fortis was gaining genes from scandens, which contributed to the sharp pointed, a more scandens like profile on fortis. Now we had followed the integration of genes from fortis into scandens and scandens into fortis using microsatellite genetics, but then we have since um collaborated with Leif Andersson who was able to look at whole genome analysis, and we asked him, we still had these blood samples. And we asked him would it be possible to find specific alleles that could have been introduced from scandens into fortis that could have contributed to more fortis point, more pointed beak? 

We know that the lead, we know that genes went in, but were there any specific ones? 

[Slide text: ALX1 a transcription factor; Photo: Group of people, including Leif Andersson, posed outside] 

So we sent our blood samples over to Leif Andersson, and Leif is here, whoops, here. And his group, and they found a very interesting gene, a transcription factor. There were others genes but one stood out as being very important, ALX1. 

[Chart: Each Darwin finch species by genotype in individual birds, with ALX1 haplotype and derived and ancestral allelic states] 

And when he looked at across all the populate, all the finch species, he found that those ones that were, had very blunt beaks like magnirostris and conirostris from Española, had the blunt form, or the blunt haplotype of this ALX1, whereas all the other finches including scandens had the sharp, the sharp beak haplotype. So he said the pointed haplotype and the blunt haplotype.

And this gene is highly conserved, you find it in fish and in mammals and as well as birds. And it’s associated with craniofacial development, and actually a mutation in the ALX1 causes cleft palate in humans. 

So what we did is we 

[Graph: Genotype (BB, BP, and PP) versus Beak shape score (degree of pointedness)] 

sent over, and also we found that in fortis, fortis had both. Now we had the measurements of all these birds, so we sent blind over to Uppsala, um the blood samples. We sent 62 blood samples over, and we found out that those that were homozygous for the pointed have ALX1 had more pointed beaks, those that were homozygous for the blunt ones had blunt beaks, and the heterozygous ones were intermediate. 

So this showed us that, and I’m cutting a long story short, but it showed us that genes from scandens, the ALX1 pointed haplotype, was moving into the fortis 

[Photos: G. scandens and G. fortis showing beak shape and length] 

population and was 

[Graph: PC2 Beak Shape by year, 1973 to 2012, with 1984 and 1985 circled] 

contributing to the sharp beaked fortis, which was now becoming more pointed, more scandens like. But the genetic exchange also goes both ways. 

[Slide text: Genetic exchange goes both ways] 

The blunt beaked haplotype was going into scandens, 

[Slide text: Average beak shape of scandens became blunter; Graph: PC2 beak shape by year for G. scandens] 

and from that point onwards the scandens was becoming blunter beaked and more fortis like. 

So, and so in both populations, in both fortis and scandens, both the genetic and morphological variation increased to a noticeable extent.

[Slide text: Genetic and Morphological variation increased; Photos: Three G. scandens showing variation in beak size and shape] 

So these three photographs were taken of scandens in 2012. The scandens at the top has a pointed beak, very similar to the scandens before introgression, before gene flow between scandens and fortis, and the two at the bottom have blunter beaks and are more fortis like. 

[Slide text: What caused this phenotypic shift?; Graph: Beak depth (mm) by year for G. fortis, Grant P.R & B.R. Grant, 40 Years of Evolution, 2014; Photo: G. fortis] 

So we can move on now to the third part. What is it that caused this huge drop where fortis became much smaller in actually body size and in beak size? So the same thing, this time it was a two and a half year drought. Where over 90% of all birds on Daphne died. [Cough] I’m sorry. So there was a, the drought but there was one other factor that was different. 

[Slide text: Colonization of Daphne by magnirostris, a close relative of fortis; Photos: G. magnirostris (~32 g) and G. fortis (~18 g)] 

That in the meantime had been colonized by magnirostris. This came in actually during the this huge El Niño event of 1983. The population gradually built up until by the time we reached 2003 there were well over 200 magnirostris on the island. Now magnirostris is a close relative of fortis. It’s an extremely dominant bird, um very aggressive. 

[Photos: G. magnirostris, G. fortis, and Tribulus seeds] 

And it loves Tribulus. So when, during this drought, when all the seeds that were left were these large, hard Tribulus seeds the magnirostris outcompeted the large fortis, leaving behind the very few, they were dying as well, but the ones that were left behind were the smallest fortis.

[Slide text: Character Displacement Event- Morphological distance between magnirostris and fortis was increased] 

So this was a character displacement event, well, and the morphological distance between magnirostris and fortis was increased during the, because magnirostris outcompeted the large fortis. So again we went back to Leif and said were there any genes underlying this character displacement event? 

[Slide text: What are the genes underlying this Character Displacement event?; Graph: Beak depth (mm) by year for G. fortis, Grant P.R & B.R. Grant, 40 Years of Evolution, 2014; Photo: G. fortis] 

And he, they found one which was again a trans—well 

[Slide text: HMGA2 a transcription factor; Photo: Group of people, including Leif Andersson, posed outside] 

it was a transcription factor facilitator HMGA2. Once again highly conserved, found in mammals, and fish, and reptiles. Now we had many of these birds banded, we had the blood samples from them before this, before they went into this two and a half year drought. So we took 70 of these, where we knew the measurements, sent them blind 

[Graph: PC (Beak size) by Genotype, Lamichhaney et al. Science 2016] 

over to Uppsala and once again asked them, asked them can you genotype this? And this HMGA2 comes in two forms. One variation was for large beaks, the other for small beaks. And once again out of the 70, the results came back that the homozygous small haplotypes, homozygous were small, those with large were large, and the heterozygous were intermediate.

So we said how many of these, taking these how many survive during the drought? Now this wasn’t just one gene, there were other genes as well, but this was the most dominant gene that we found. 

[Slide text: G. fortis Survival according to HMGA2 genotype 

Genotype Alive birds (n) Dead birds (n) Survival ± SE (%) 

LL      6        14         30.0 ± 10.2 

LS     17       15          53.1 ± 8.8 

SS     14        5           73.7 ± 10.1] 

And so out of these, the large haplotypes, out of the twenty birds fourteen of them died. Whereas out the small haplotypes, out of 19 birds only five of them died. So the 

[Slide text: The HMGA2-L haplotype associated with large beak size was at a strong selective advantage (selection coefficient s = 0.59) facilitating this phenotypic shift; Graph: Beak depth (mm) by year for G. fortis, Grant P.R. & B.R. Grant, 40 Years of Evolution, 2014; Photo: G. fortis] 

the HMGA2 large haplotype which was associated with large beak size was at a strong selective advantage throwing the, through this selection event. And it facilitated this phenotypic shift. 

Now not only was there a morphological change that occurred, but quite unexpectedly 

[Slide text: Behavioral influence 

G. magnirostris has a loud song sung in the same frequency band width as fortis and scandens songs; Photos: G. magnirostris and G. fortis] 

there was also a behavioral change. Now to tell you a little bit about the the song, magnirostris has a loud song sung in the same frequency band width as fortis and scandens songs. 

[Slide text: 8 fortis and 2 scandens learned and sang a magnirostris song 

None bred with magnirostris. Why?; Images: Spectrograms of G. fortis, G. scandens, and G. magnirostris songs, and of G. fortis singing G. magnirostris song, and G. scandens singing G. magnirostris song] 

So this is what a magnirostris song looks like. Sometimes they’re two notes, sometimes three notes. And over the forty years study we’ve had eight fortis and two scandens that learnt and sang a magnirostris song. So here’s a normal fortis song, and this is one of the fortis that sang a magnirostris song. A normal scandens song, and one of the scandens that sang a magnirostris song. 

Now none of these birds bred with a magnirostris. So why was this? The reason was that when these poor little birds opened their mouths, sang a magnirostris song, a magnirostris which was twice its size whipped in as if from, as if from nowhere, and just beat it to bits. 

[Laughter] 

So so it never got anywhere. So it seems that 

[Slide text: When the size difference between species is large the barrier is robust.

Even learning a magnirostris song did not lead to mating; Photos: G. fortis (~18 g) and G. magnirostris (~32 g)] 

when the size difference between species is large, the barrier to reproduction is robust. Even learning a magnirostris song didn’t lead to mating. But then we, an extraordinary thing happened, and we almost could hardly believe our ears when it happened. 

[Graphs: Number of notes per second versus Frequency band width for G. fortis, G. scandens, and G. magnirostris before and after magnirostris colonization] 

But the songs of fortis and scandens after magnirostris had really built up began to change. So this is before, when there were only very few, I think there were 1 2 3 4 5, magnirostris is in green, magnirostris males on the island. This is what the scandens songs up here in red, and fortis songs like like look like in acoustical space. 

And then after magnirostris built up until there were well over 200 of them on the island, then these birds both scandens and fortis began to sing more notes per second. So that it was almost as though the song was compressed. 

[Images: Spectrograms from before and after magnirostris colonization] 

So instead of getting songs like this, we got more songs like this. And as I say we could almost hardly believe our ears and we started to wonder had it been a change in the equipment we used, so we went back and re-analyzed all our songs from before. And no it was truly a change that we were hearing. And so we wondered why there had been this change. 

[Slide text: -Podos, Nature (2001): Beak size and shape influence song production -Competition for acoustical space can result in song divergence 

Blue-tits songs changed in presence of Great Tits Doutrelant & Lambrechts (2001) 

Ant bird songs are more divergent in sympatry than in allopatry. Seddan (2005)] 

Now Jeff Podos had written this very influential one about beak size and shape influence song production, so we thought perhaps this is what we’re seeing. But we analyzed our work our work, and we saw absolutely no association between beak size and shape and the song production. So clearly it wasn’t this that was going on. 

And then we wondered where, we started various things, but then we came down to could it have been competition for acoustical space that resulted in song divergence? People have shown this in blue-tits and in and in other birds such as the ant birds. And there have been several papers that have shown this change in acoustical acoustical space as a result of a bird entering a new environment. 

[Slide text: Did the change in songs occur in adults or during the process of learning?; Graphs: Number of notes per second versus Frequency band width for G. fortis, G. scandens, and G. magnirostris before and after magnirostris colonization] 

And we, so we wondered if this change had occurred as adults, but it hadn’t. The adults had not changed their song, we recorded them every year. And it, the song change occurred during the 

[Slide text: Copying of father’s song shown in the significant regression line in red R2 = 0.54 and 0.43 

Solid points above the slope of one indicate that sons sang more notes per second than their father Wilcoxen matched pair P = 0.003 fortis, P = 0.03 scandens Grant B.R. & Grant P.R. Nov 2010 PNAS; Graphs: Father trill rate versus Son trill rate for G. fortis and G. scandens] 

short sensitive period of learning. So we looked at the father’s son and we found that the copying of father’s song was still, they still, the son still copied the father’s song. And you can see this by the red regression line. But if you look at the solid points above a slope of one, it is an indication that sung, sons sang more notes than their father. So this seems to have been what was going on, and we saw this both in fortis and in scandens. 

So this was really very interesting to us because it answered or helped to answer another question that we had had for many years which is 

[Slide text: Competition for acoustical space- Acoustical Interference; Images: Spectrograms of G. fortis and G. scandens songs from Daphne, Santiago, Pinta, and Santa Cruz] 

why on all the different islands in the Galapagos, um fortis and scandens and other birds were singing different songs. So for example this is fortis song on Daphne, scandens, and scandens song on Daphne, but on Santiago, Pinta, and Santa Cruz for example, they, fortis is still singing a modulated song but with completely different notes. 

Scandens still sings repeated note song but with completely different notes. And on any of these islands if you look at the acoustical space, then all the birds, I mean for example there are over nine species of finches on Santa Cruz, but they all sing in their different acoustical space. So they’re separated acoustically on the different islands just as they were in, on Daphne. So this gives a clue that there must be some interference in acoustical interference. 

[Slide text: Origin of a new lineage; Photo: Isla Daphne] 

So now move on to the origin of the new lineage. Now we had already shown with introgression between fortis and scandens populations that there had been 

[Slide text: -Introgression between fortis and scandens populations increased the morphological and genetic variation -In a new environment this could lead to evolutionary change along a different trajectory] 

this increase of morphological genetic variation, and we had actually written having found this, that in a new environment this could lead to evolutionary change along a different trajectory because of the huge increase in both morphological and genetic variation. But we never ever dreamt that we could see such a thing. 

[Slide text: Unique song, Large 28.5g; Photo: Large finch; Image: Spectrogram of song] 

So but what happened is that a bird arrived on Daphne at the time when all the birds on the island were banded, and Trevor Price was a graduate student at this time. And he caught the bird and he measured it, and it was a large bird. It was twenty-eight point five grams whereas a normal fortis is 17 grams. So it was a giant. And when it, it was a young bird when it arrived, and then when it started to sing as an older bird it had a completely unique song, never before heard on Daphne. 

[Slide text: 5110 -FSF Hybrid backcross from Santa Cruz?; Photo: Large finch, 5110]

Now at this time we had microsatellites, and we took, Peter and I went back to the island, we took a blood sample from this bird and we matched it up with every bird on Daphne, and clearly it was not born in Daphne, couldn’t possibly have been this. But it came out with 95% confidence that it was, in an assignment test, that it was a fortis–scandens–fortis backcross born on Santa Cruz, the nearest big island. 

This made quite a bit of sense, I mean the island is only five miles away and it could possibly have flown over from Daphne. But soon we published this. How wrong we were because when with Leif’s group, when they were able to look at the whole genome they found that this bird was 

[Slide text: Lamichhaney et al. 2018. Science 359, 224-228; Images: G. conirostris and map of Galapagos Islands] 

a conirostris from Española. Now Española is yeah, well over a hundred kilometers away but it came out as a perfect conirostris. So what what it did is that it took a long time to breed, but when it did breed 

[Slide text: All offspring had genetic transmission from 5110 

All males sang 5110 song; Image: Family tree of Line B] 

it bred with a fortis–scandens–fortis backcross born on Daphne. And that this to us at the time made perfect sense because we thought it was a fortis–scandens–fortis backcross on Santa Cruz. And again we were completely wrong. It produced a few off, offspring but none of them survived. 

Then it bred eventually with a fortis and produced some offspring which did survive. And this was the only outbreeding that occurred. We followed all these offspring, we took blood samples from them. All of them showed genetic transmission from 5110, and this was seen from microsatellites. 

All males sang a five, the 5110 song, which was… I should say we banded this bird 5110. All males sang his song, which had been passed down from father to son. And so we were following this, and then along came this two and a half year drought, when 90% of the birds died. All these birds went out, except for these two, an inbred brother and sister. 

When the rains came back 

[Slide text: All birds showed genetic transmission from 5110 

All males sang 5110 song 

All were large; Image: Family tree of Big Bird Lineage] 

this inbred brother and sister bred with each other and produced 26 offspring. All but nine of them survived. We had a daughter breeding with her father, a son breeding with his mother, and the rest of the sibs breeding with each other to produce more offspring, which bred with each other to produce more offspring, which bred with each other. 

All males, all birds showed genetic transmission from 5110. And now we have complete genome sequencing of all these birds and they all show an absolute genetic transmission from 5110. All, they all sang 5110 song, and all were large. 

And of course the inbreeding 

[Graph: Genome-wide inbreeding coefficient (F) by generation] 

from whole genome analysis increased each generation, from generation one up. When they first started there was quite a lot of heterozygosity because it was a hybrid. And then as they were mating, with the brother and sister mating, and all the offspring the inbreeding got more and more intense. 

[Graph: Allometric shift- PC (Body size) versus PC (Bill size) of G. conirostris, Big Birds, and G. fortis] 

Now what also happened was that there was an allometric shift. So this is fortis in green, conirostris in red, and this is what we call the Big Bird lineage from 5110 is displaced in this allometric shift. Now this is really interesting because it has a small hea—, it has a small body, it has a large beak. So it has a large head or large beak on a small body. 

And this makes it really efficient at feeding on Tribulus. It is absolutely omnivorous, it feeds on everything that fortis and scandens on the island feeds on. 

[Graph: Beak Depth (mm) versus Efficiency for G. fortis, G. magnirostris, and Big Birds] 

But it’s particularly efficient on this large, hard Tribulus seed. It’s just as efficient as magnirostris, but being smaller it requires fewer seeds to survive. So it might, if there’s another drought, and Tribulus is abundant in another drought it could do very well. So is this new lineage behaving like a 

[Slide text: Is the new lineage behaving like a separate species? Large; Photos: G. fortis and 5110] 

separate species? It’s much larger than its nearest relative fortis. 

[Slide text: Differ in size; Graph: Beak Depth, mm versus Beak Length, mm for G. fortis, G. magnirostris, and Big Birds] 

In morphological space it lies between magnirostris and fortis. And this was the gap that increased when magnirostris outcompeted the large fortis during the character displacement event I told you about. And the Big Bird population moved right in. 

[Slide text: Differ in song from all other species on Daphne; Images: Spectrograms of songs of three generations of Big Bird, normal fortis, scandens, and magnirostris songs, and photos of finches] 

This is three generations of the Big Bird song. And I should, and this is the normal fortis song, normal scandens song, and a normal magnirostris song. 

[Slide text: Pairs hold contiguous territories 

Overlap with G. fortis, G. scandens, and G. magnirostris; Photo: Isla Daphne with colored dots indicating different territories] 

It breeds in one part of the island. Irby will recognize this part of the island. The blue dots are the area before the drought, the red dots after the drought. They hold territories that are contiguous with each other, which they defend against each other. But these territories are very large and they overlap with the territories of fortis, scandens, and magnirostris who they ignore, and who are ignored by them. 

[Slide text: Thus the new lineage is functioning as a separate species; Photo: 5110] 

So in all respects the newly lineage is functioning as a separate species. Will it die out 

[Slide text: Will it die out through inbreeding depression? Will the genetic variation be augmented through gene exchange?; Photo: 5110]

through inbreeding depression? Well you can see it’s quite highly inbred, but it’s still not completely inbred. And we see no sign of inbreeding depression so far, but of course it might die out because of that. Will the genetic variation be augmented through genetic exchange? It might be, but so far we’ve not seen any sign of that yet. 

[Slide text: Whether it survives or not this new lineage gives insights into how a new species could arise and either persist or becomes extinct.

“—to understand the mechanism of speciation, the focus should be on cases of incipient speciation rather than on completed ones” A.W. Nolte, D. Tautz Trends in Genetics. 26, 54-58 (2010)] 

So whether it survives or not, this new lineage gives us insight into how a new species could arise and either persist or become extinct. 

[Slide text: Summary; Graphs: Beak Length versus Beak Depth of G. fortis and G. scandens in 1973 and G. fortis, G. scandens, G. magnirostris, and Big Birds in 2012] 

So in summary, we went to the island when there were two distinct species, fortis and scandens. And so after the, this enormous El Niño event of 1983, largest in 400 years, then the hybrids began to survive and fortis and scandens began to converge on each other. Magnirostris came in, outcompeted the largest fortis, and the Big Bird lineage moved right in here. So this gives us 

[Slide text: Insights into 3 models of speciation 

Allopatric model (fortis & magnirostris) -No interbreeding on secondary contact -Character displacement led to morphological divergence -Competition for acoustical space led to song divergence] 

insight into three models of speciation. There’s the allopatric model of speciation, the kind that really has been called this, but Darwin actually first envisaged this. And this we saw between fortis and magnirostris where there was no interbreeding on secondary contact. A character displacement event led to the morphological divergence, and we saw competition both for acoustical, morphological space and for acoustical space, which led to song divergence. 

[Slide text: Under appreciated role of introgression -Speciation by fusion- Interbreeding led to one unique lineage of mixed genetic composition (fortis & scandens) -Speciation by Introgression- In a new environment led to development of a unique lineage (BB lineage)] 

But there’s this under appreciated role of introgression, gene flow between two different species. And we had speciation by, could lead to speciation by fusion where interbreeding between fortis and scandens on Daphne led to one unique lineage of mixed genetic composition very different from fortis and scandens on other islands in the archipelago. 

And then speciation by introgression with conirostris, a conirostris coming in to Daphne. And in a completely new environment led to the development of a unique lineage, which we call the Big Bird lineage. 

[Slide text: Darwin: Notebook B. 1837; Image: Diagram from Darwin’s notebook] 

So Darwin, I think we all know this diagram. It’s his, Darwin’s only diagram, which was in his notebooks where he was pondering over the relationship between species, and drew this sort of tree-like diagram, and wrote “I think”. 

[Slide text: … I now realize; Image: Darwin’s diagram with “I think” written at the top] 

Would he now join up some of these branches, and write “I now realize”? I don’t know. 

[Slide text: We thank: -Many Graduates, Post-doctoral Fellows, Collaborators and Daughters -The Galápagos National Parks and Charles Darwin Research Station for logistical support -McGill University, NSERC, NSF, the Balzan and Kyoto Prize Foundations; Photo: Isla Daphne] 

But we thank many graduates, post-doctoral fellows, our collaborators, and daughters, and one of these we thank very much, Irby, who spent six month’s hard work on this island. The Galapagos National Parks and Charles Darwin’s Station for logistic research and our funding agencies. 

But anybody, and I’m sure I include Irby in this too, anybody who has worked on the Galapagos I’m sure we want to leave you with two messages. One a conservation one, that environments and populations are dynamic, they’re constantly changing, and for a sustainable environment, or a sustainable globe, we must keep them both capable of further natural change. 

The other one, and I think this is directed to the young people in the audience, you live, we live in very exciting times. The genomics data are rapidly accumulating, changes are happening every year. And this can really enhance our field studies. But reciprocally a reliable interpretation of genetic data really requires a deep understanding of ecology, evolution, and behavior in the natural environment. So I think it is very important to remember this. Thank you very much. 

[Applause] 

[Rosemary] Absolutely, yes, yes. Yes? 

[Audience question] 

[Rosemary] Oh, okay. You know I almost started showing a little video about this because we’ve got to be very careful when we go on, because, well it requires quite a lot of permits, but also we nowadays we have to go through a quarantine process. So we have to make sure that everything we take on is scrupulously clean, all the food that we take on has, so that goes into quarantine. 

All the food that we take on has got to be checked, so for example with things like rice and flour we freeze them for 96 hours before we take them on. And everything is sprayed and inspected and put into large bags, and taken onto the island then. 

I think this has changed a little bit since Irby went, but we were partly responsible for making this very, we were talking to the parks to make it like this. And they now do this for all all islands in the Galapagos. 

[Audience question] 

[Rosemary] Yeah, yep. That’s a super question. We have also looked at this a little bit, and we found evidence of this in one plant, which is Cordia lutea, which we, and we have published this. But also I think what you’re particularly referring to which we’ve also thought about is the Tribulus. How does, has that actually increased in size? 

Now Mark Johnson is looking at this right now. And he’s looking at the, to see whether the Tribulus have changed at all. We started to do this, but we thought, it was. It got a little bit difficult because it’s it’s a perennial plant. It grows in different parts of the island, and it is extremely variable. And I think a lot of variation is due to just the soil it is in and the amount of water it gets at that time. 

But that, but Mark is actually looking at this really thoroughly now, and so, and looking at it across many populations in many different islands, and actually in different parts of the world. So so we you know so there might be change. And we certainly in Cordia lutea between the population on, that occurs on Genovesa, and also on Española. Then those seeds are, the size of the seeds are, very much reflect the bird’s beak. 

So in magnirostris on Genovesa they take the bottom third of Cordia lutea on that, and the Cordia lutea has a great big seed like a cherry stone, a big cherry stone there. On Española the Cordia lutea, exactly the same species, it’s much smaller seed. But there conirostris takes it, and the conirostris is a smaller bird, but it also takes the bottom third. So it looks as though there has been selection on the seeds. 

[Laughter] 

Yeah, yeah this question up here. 

[Audience question] 

[Rosemary] Yes? Yes? We, yes what I’ve just told you about is as you say the external phenotype. I, what we hope to with the Uppsala group’s help. It may be trying to find out some of these, but there must be some really strong physiological changes in some species of finches. 

For example there’s difficilis, which is a sharp beak finch. There’s populations that live high up in the sort of cloud forest. And they they must be very different from the ones that live lower down in the arid climate. And it would be really nice to know the physiological differences there. 

You’ll see the same in the warbler finch. There are two warbler finches, Certhidea. One of them, and they’re genetically very different from each other, they look identical but they’re genetically very different. One of them only lives high up, the other in arid, in the arid part of the environment. And it would be really nice to know the physiological differences there. And there, I could go on and on, but yes, you’re absolutely right, and we hope, if not us, because we’re really quite old now 

[Laughter] 

but but if we could live another many years that would be what we would look at. And we are starting to do this with Leif, mmm. That’s a great question. 

Yeah? Yes. 

[Audience question] 

[Rosemary] Yes, we’ve we’ve looked at genetic, we’ve looked at the behavioral differences, we’ve looked at the genetic differences, and we, and there will be, we’re trying to actually get even longer reads of genotypes so that we can actually see if there are any things like inversions going on like this. So that again is work in progress, yeah. 

[Audience question] 

[Rosemary] Okay, yes. I mean there have been long periods of time when when I’ve shown you this where they’re all the same. I mean that, I just showed you beak depth was all the same. But actually you know by living on the island with, there are constant questions every single day. The, there are many things that I haven’t talked about of course because you know I had an hour to talk, and so I had to pare down on everything. 

But it’s it’s a very exciting place to to be, and when you’re actually living amongst the birds yourself, you’re just bombarded with questions every day. Even if it’s not the birds itself, it’s questions like the one I got over here is are the plants changing? 

So we’ve done other little experiments about pollination experiments on Opuntia and things like this. But there is so much going on that you’re just bombarded with questions. And so it’s a very stimulating place to be, and I can honestly say, and I’m talking for Peter as well I’m sure, that there has never been a moment when we have been bored or or 

[Laughter] 

and in fact we’ve taken books down and not read them. You get, you get really exhausted because you have you know 12 hours of light only. And you’ll, and it’s amazing how long you know, you’re so exhausted because of course it’s very hot, that you’ll you will sleep very well at night even though it’s on a rocky, rocky slope. The only thing I will say that is nice when you finally get back is that that first fresh water shower is just wonderful 

[Laughter] 

Yeah, but otherwise, no, you never get bored. Yeah.

[Applause] 

[Rosemary] Thank you very much. Thank you.

End of transcript