You are here


  • Post date: 1 year 8 months ago
    Citation for this post: BibTeX | RIS

    In 1988,   published showing that male barn swallows (Hirundo rustica) with experimentally-lengthened tails obtained mates quicker that males with shorter tails. While this wasn't the first experimental demonstration of the function of male secondary sexual ornaments, what made this study important was that it was the first in a monogamous species. Twenty-eight years after it was published, I spoke to Anders Møller about his motivation for doing this study, and what we have learnt about mate choice in barn swallows since then. 

    Hari Sridhar: This is not your first piece of work on barn swallows. You had, already at this time, done quite a lot of work on this species. Can you tell us how you got interested in them?

    Anders Møller: Well, in actual fact, this year is the 46th year when I am studying these birds. It’s, almost, an embarrassingly long time I have dedicated to this species. But then, as you know, people in biology study all kinds of organisms for very long periods of time – Drosophila in animals, and for people who are studying plants, it is Arabidiopsis. I started getting interested in the behaviour of animals – free-living animals, outside the windows, in the real world - many many years ago. When I was studying biology at university, I was a little bit frustrated that there were so many people studying animals and plants in the lab, under conditions that might not have very much to do with the conditions under which they live in nature. So the motivation really was to try to understand how animals behave in the free world.

    I have studied many different kinds of organisms over the years, because I can’t make up my mind. Different kinds of organisms are suitable for different kinds of questions. The reason why I study barn swallows, and why many other people have joined this effort, is because they are extremely abundant, very easy to handle, very easy to observe and very easy to catch. This might not seem like important advantages to a non-biologist, but, believe me, there are very few organisms where you can catch individuals and follow them throughout their lives. There are many organisms, including Drosophila, where you can’t do this. They might be the wrong size, or if they are the right size, you never see them again once you have captured and tagged them. In summary, I think you can even call it laziness, to choose a species that is easy to study. But if you want to get robust results you have to pick a model system that is easy to work on.


    HS: Right from the beginning, was understanding the reproductive behaviour of this species one of your interests?

    AM: Yes, already when I was a master’s student I was studying these birds. I grew up in the countryside. My father was a farmer and I was also a farmer for several years. And in the countryside there were swallows everywhere. In Europe, as well as in North America and Asia, it’s a species that signals the arrival of spring. There are many superstitions about this bird - if you don’t take care of this bird you can have problems of running your farm, your animals might die and so on. All these kinds of superstition.


    HS: When did you get interested in sexual selection?

    AM: This was around 1985-86, when I had just finished my PhD and obtained an assistant professorship at the University of Uppsala in Sweden. The way I looked at sexual selection in those days was very influenced by what Charles Darwin had written about it . He emphasized that in species where males could attract one or more females, the males that were more adorned would have, on average, more mates and produce more offspring. This is easy to understand for a species like the peacock, but for a species like the barn swallow, which is socially monogamous, it’s more difficult to understand. If there is an even sex ratio in the population, which is usually the case, then there is a partner for everybody.  There is no variation in mating success and therefore you shouldn’t expect any traits to be exaggerated. If you look at barn swallows, with a few exceptions, all morphological characters are the same size in males and females. One of these exceptions is the length of the outermost tail feathers, which is considerably longer in males than in females, and also much more variable. The puzzle was that you have this exaggerated character but there appears not be to be much variation in mating success. How can this come about?


    HS: What was your PhD dissertation on?

    AM: I wrote a thesis on the costs and benefits of social behaviour in barn swallows. If you start studying this species you will notice that quite a few individuals are completely solitary - only one male and one female breeding in a particular site - and in some places you get 200-300 pairs breeding in one single barn. I was interested in finding out why there is this huge variation in sociality and what are the costs and benefits associated with that kind of behaviour.


    HS: In the abstract of your paper you cite five references, four of which are theoretical papers. The fifth is . Can you tell us about the influence of that paper on your work?  

    AM: Yes, I can easily do that. As I said before, Charles Darwin said that in many species you can easily have exaggerated secondary sexual characters - exaggerated morphological traits, vocalizations, smells - because there is so much variation in male success. This was Charles Darwin’s intuition that led to this conclusion. Subsequently, there was not much done in this area. That’s really a strange feature of sexual selection - for decades and decades and decades nothing was done in terms of empirical research. And , at the University of Gothenburg in Sweden, was the one who changed that. He had been to Kenya and he had seen widowbirds. Male widowbirds are almost as exaggerated as peacocks. They have extremely long tail feathers, which they display in flight. The females are smaller, inconspicuous looking birds, which take care of the offspring and build their nests in grass. These females, according to Darwin’s theory, should be most interested in males with the most exaggerated traits. What Malte Andersson did was to simply bring a pair of scissors and superglue and, by cutting off a piece of the tail feathers from some individuals and adding this to the tails of other individuals, change the variation in exaggeration of these traits. The prediction was very simple – you should have more females aggregating on territories of males with experimentally elongated feathers. This was what Malte Andersson was able to show, subsequently.

    However, there were still some outstanding questions. And one of them was that if you go out and look at the natural variation in tail length of these widowbirds, you should find that individuals with naturally longer tails should be more attractive to females. But Malte Andersson was not able to show that in his field study. So, this was really a puzzle. You could manipulate these feathers and that would have a consequence for mating success of males, but then if you look at natural variation in length of males, there was no effect of tail length whatsoever. Malte Andersson writes about this, in one of the final notes in his paper, but it was still hanging in the air and it was kind of puzzling why there was this inconsistency in the results.


    HS: Was your study aimed at examining this further?

    AM: Yes, I had two objectives. One of them was to test what could be the mechanism behind the evolution and the maintenance of exaggerated secondary sexual characters in monogamous species. One solution to this problem was proposed by Ronald Fisher, if it is the case that some individuals are in better condition than others – and there are many reasons for expecting that to be the case – then you would have some individuals that breed early and some that breed late. The ones that breed early should be the ones that are most attractive in the population. And therefore, for a monogamous species, one would expect an association between the degree of exaggeration of these secondary sexual characters and timing of breeding. So if you manipulate tail length the timing of breeding would change. The second objective was to measure the lengths of the outermost tail feathers and investigate if natural variation in this character was associated with timing of breeding. So that’s what I set out to do and I did it.  Of course, it’s much easier again to do this on barn swallows than on widowbirds. Malte Andersson did this in the field in Kenya where he was always surrounded by perhaps 200 locals, 75 to 100 of whom were screaming children. It was not an easy task. He doesn’t mention this in his paper, but he told me subsequently. I had a much easier deal. I rarely had more than one person attending these events. It was also easier in another respect - the barn swallow has only two exaggerated feathers so it’s much easier to make this manipulation.


    HS: In the Acknowledgements, you also thank Malte Andersson for some advice on the glue you used to stick the feathers. Can you share with us what this was?

    AM: Well, the glue we used was simply a super glue. Today you can buy this in any store anywhere in the world. Its glue that basically turns very hard as soon as it interacts with the air and so the feathers will never break in that particular spot. Sometimes feathers break in other places but that’s another story. I think like with most of these inventions, this glue was something that was invented by the military industry in the US. The brand we used was, I think, called Loctite.


    HS: Was your sampling design different from Andersson’s?

    AM: Well, in fact, if you read Andersson’s paper carefully, you will see that he really doesn’t measure reproductive success as the number of females that each male could attract. He went out to the field and recorded the number of nests in the territories of the different males. But if you have ever been to the savannah in Africa, you would have noticed that this is not an easy task to do. The grass is very tall and the nests are camouflaged and inconspicuous. And because of this it is very difficult to record accurately. Also, in Andersson’s design, he recorded the number of nests before he did the experiment, then he cut and glued the tail feathers and then, after some time, he went back to recount the number of nests in the territories. So his measure of success was the number of nests after minus the number of nests before. What I did was to manipulate the tail lengths of males and then record when the females were attracted to the territory. But I also recorded the exact date when these females started laying their eggs. So, in that way, I could show that if you elongate the tail feathers, the males and female partners started breeding earlier, because the females in better condition were attracted to these males with elongated tails. This has a strong consequence for reproductive success because, not only in swallows, but virtually in all different kinds of animals, those that reproduce earlier are more successful in reproduction.  In seasonal environments, at the end of summer if the offspring are not fully independent they will die. So you can only achieve high reproductive success by starting out early. In the barn swallow case the birds can produce from one to 3 clutches per year. But you can only do that if you start out early. For that reason, the males with elongated tails reproduce much earlier than those with shortened tails and this resulted in a more than two-fold difference in reproductive success.


    HS: You did these experiments in a place called Kraghede in Denmark. What kind of a place was this? Do you continue to work there?

    AM: Yes, this is the place where I grew up, so I know all the farmers in these farms since I was a small boy. In fact now I think none of those farmers who were there when I started this study are still around. Many years have passed and most of them have passed away.

    Kraghede is not a particularly interesting place. It’s an open country site with a few farms and few trees around the farms. The swallows, as I said, breed mainly inside buildings and barns. Therefore they are very easy to catch. I catch them using mist nets. You can simply put the nets in the doors and windows and eventually you will catch all of them. As I said before, there is this advantage with this bird that, not only can you catch all of them, but if you are interested in measuring long-term consequences of these kinds of experimental manipulations, you can also catch all the survivors in subsequent years, because they return to the site where they were produced.


    HS: Did you do all this work on your own or did you have people to help you?

    AM: I often had some of my friends joining me, but usually, after some weeks, most people think it is too boring. In fact, for many years, I have always brought my children with me to the field. I remember, one morning some years ago, after we had been in the field for, I think, six or seven weeks in a row, my son Pierre standing outside the car and saying “Dad, I don’t want to see any more swallows!”


    HS: How old was he?

    AM: I think he was 4.


    HS: What was being grown in these farms?

    AM: Well, in the 70s and 80s it was mixed crops - grain and grass and hay for the cows and so on. But nowadays it is mostly wheat all over.


    HS: Do you continue to work in Kraghede?

    AM: I have just come back to Paris after spending six weeks in that site. I have just captured all the birds, found out where they were living , who has survived from last year, how many eggs they have laid, how many young they produced during the first clutch.


    HS: 28 years after this study!

    AM: Yes, and 46 years since I first started working there.


    HS: Has the site changed a lot since you worked there in 1987?

    AM: Yes, because in those days it was much more diverse and now it’s a monoculture. Just one or two crops everywhere. The reason why the swallows like the farms and the farm buildings is that they eat the flies that live in the surroundings of the farms. But now, because most farmers have given up keeping cattle, there are many fewer insects for the swallows to eat.


    HS: Have the arrival dates of these swallows changed since the time you did this study?

    AM: In my study area, the birds now arrive, perhaps, 2 - 2.5 weeks earlier in spring, than some years ago. This is also one of the reasons why I am still working on these birds. There are very few long-term studies that go beyond 10-20 years. and found that the effects are quite strong. Like I said earlier, males with long tails attract good-condition females earlier in the season, breed earlier and have more time to sire a second clutch. But with the advancing of spring arrival, this advantage for long-tailed males is not so big anymore, because even the short-tailed males now have time for a second clutch.


    HS: You have also made some behavioural observations in the study, in the context of “female rejection”? What time of the day were those observations made?

    AM: Well, this is the other reason why most people are not very keen to join this fieldwork. I usually start very early in the morning. At sunrise. And in northern latitudes like Denmark, it starts getting light very early – around 3.30-4 - in the morning. And that’s the time when the birds are most active. The males are displaying their tails to the females, to attract them to their small territories, where they eventually will construct the nest.


    HS: In a couple of places in the paper, you mention a “manuscript in preparation” One is where you say “More than 90% of all pairs remained together during both first and second clutches in the same breeding season, whereas mate fidelity between years was low, primarily due to high overwinter mortality” and the other is where you say “Male tail length was not related to territory size nor to his investment in nest building, mate guarding or the feeding of nestlings.” When and where were these eventually published?

    AM: They were published in So if you take the last point first, Darwin was puzzled by sexual selection in many different ways. He wrote explicitly about the peacock, that it was a bird that kept him awake many many nights, puzzling about how can something like that evolve and be maintained. He felt the peacock was the largest objection to this theory of evolution. This was simply because he had difficulty understanding why should it matter so much to these females if they were associated with one male or another. So the small note that you just read was one of the ideas that Darwin thought about. He felt that perhaps males with more exaggerated traits were better fathers, who provided more food for the offspring. But that is not the case in the peacock, because the peacock after copulating with the female never sees the offspring. He doesn’t feed them at all. In the barn swallow it’s a little bit different. Males contribute approx. half of the food to the offspring. About the note that you just mentioned - I followed the birds and recorded how often they fed their offspring. It turns out that, in fact, it’s the long-tailed males, the attractive males, who are the lazy fathers. They provide less food, not more food, which makes it even more puzzling. Why do females accept this cost?

    Subsequently, I also looked at other explanations. One of them was the explanation put forward by , where they suggested that what could be running these circles of sexual selection in all these different species could be that some individuals are more resistant to deleterious diseases and parasites. If that is the case, it is not difficult to imagine that such males with exaggerated traits would be healthier and, for that reason, females would be out looking for such males because they will also gain this advantage.


    HS: Just is an aside: In one place in the paper there is a reference to citation no. 22, but the paper has only 21 citations.

    AM: Maybe it is a citation in one of the figures, because usually those come at the end.


    HS: But this one comes in the middle of the paper, so maybe it’s a typo. It comes where you say “It has been suggested that sexual selection may occur in apparently monogamous systems if females engage in extra-pair bond copulations with favoured males”?

    AM: Let me just check this. Yes, that was the other issue that came up much after Darwin and Fisher. Darwin had a rather Victorian view of nature and relationships.  He thought that females would always be monogamous. But that is not necessarily the case. In fact, Table 1 in the paper, and also our subsequent DNA fingerprinting studies (, ), showed that individuals with lengthened tails were more attractive in two ways: not only did they have more offspring in their own nest, they also fathered more offspring in the nests of other individuals. Because of this you need to look at variation in number of partners as an additional mechanism that would generate variation in mating success.


    HS: How long did the writing of this paper take?

    AM: I am an old man now and soon I will retire. I have come to the conclusion that one of the most important skills for a young scientist is to be able to write clearly and precisely. Surprisingly, this skill is usually not taught in universities. Or if it’s taught, it is only because of the initiative of individual supervisors who try to help their students.  What I have done in the last few years, not only here in France but also in Spain, Italy and China, is to give courses to students on how to write papers in an efficient way. And my take on this is that if you understand the questions and understand the results, you can write the paper in one day! You can look at this in a different way.  All journalists have a deadline. They have to quickly write something that interests the readers of their newspaper, otherwise they will get kicked out by the editor of the newspaper. I can’t see much difference between writing a paper or an article for a newspaper. So I usually recommend to students that they put down as much as possible in one single day. Today, with all the software and computers, you can start editing immediately, and after a few days or a week you will have made very good progress. And it is not just something I say. When I give these courses I take out my computer and start writing a paper in front of the students and finish in less than one day. I do that to show them that it’s possible. You can teach many things, but if you can show it with a good example it’s so much more powerful.

    This paper too, took very little time, because I had been thinking about these questions for a long time and had already collected all the data. It was just a question of putting it down. As you know, the paper is not very long, so it was more a question of trimming it and cutting off small pieces that were perhaps not so relevant. I think I typed everything for this paper in one day. Of course, this first version is quite different from the published version. But you have to start somewhere. I know many students who, when they start writing their paper, start with the title, then write an abstract, but then they think the title is not very good, so they come up with another title. In this way, after an hour or two, they are still at the title stage. So, I usually try to put everything down immediately and not think much about whether it’s perfect or not.  That’s something you will have to improve subsequently.


    HS: When I start writing a paper, I have already decided that it is going to take a few months at least, so I am always working with that timeline at the back of my mind.

    AM: I don’t look at it that way. Many people say that scientists and artists are two different kinds of human beings. I don’t agree. If you have done a study that is very close to your heart, you want to get it out on a piece of paper and show it to everybody as quickly as possible. This is not so different from the feeling of an artiste who is trying to write a poem. I usually say to my students when I give this kind of course: you should feel that if you can’t get it out now you will die!


    HS: Did this paper have a smooth journey through peer-review?

    A: Nothing is easy. It is always very hard to publish papers. You need to be working on the right subject at the right time. You need the right referees who are not too ruthless. I always say that it is possible to reject any paper. Any paper! If you have three positive reviews and the fourth is extremely critical, it’s done. You can do whatever you want and it will not change anything. You will have to look to another journal.


    HS: But this paper seems to have had a fairly easy ride: it was submitted in December 1987 and accepted in two months.

    AM: Yes, but I think it went through two rounds of review.


    HS: Did it change a lot from the first draft?

    AM: Not really, just some details needed to be specified. In general, the reviews were very favourable. The fact that the study was experimental, that it addressed this old question of Darwin and Fisher, and had these interesting results in Table 1 on extra-pair copulations – all of these made the reviewers think that it was a reasonably large step forward in this field.


    HS: In the Acknowledgements you thank someone called A. Ulfstrand for the drawings. Was this for the numerical figures or the drawings of the swallows?

    AM: Both. In the old days there were no software, so all the illustrations were made by hand. A. Ulfstrand was the wife of the head of our department. In fact, I am still in contact with my former head of the department and his wife. I still see them every now and then and we talk about swallows and the old days.


    HS: Do you still have the original drawings?

    AM: Yes, I still have them in a folder in my office.


    HS: Can you tell us about the other people you acknowledge in the paper?

    AM: , and Arne Lundberg were colleagues from my department in Uppsala. was the head of the department. Jonathan Waage was a US scientist on sabbatical in Uppsala in those days. was a colleague of mine; we were collaborating on many different projects in those days, so it was natural that I asked him. was another North American scientist who was on sabbatical in Uppsala at that time. was a theoretical evolutionary biologist from the UK.


    HS: Was the grant you acknowledge awarded, specifically, to study sexual selection in barn swallows?

    AM: Yes, it was.


    HS: When the paper was published, did it attract a lot of attention, within academia and in the popular press?

    AM: Well, this has changed a lot since those days. Today if you publish a paper like this, you will have lots of people contacting you. In those days that was far from the case. There was a “News and Views” piece in Nature about the paper by Peter O’Donald.


    HS: This paper has been cited so many times. At the time when it was published did you anticipate that it would have such a big impact and become such an important study in the history of sexual selection?

    AM:  No, certainly not. I think there are very few people who do research in order to attract attention.  I think most scientists do it because they want to know, want to satisfy their curiosity. At least to me, personally, it doesn’t matter if it’s cited or not. This perhaps sounds a little bit odd, but the most important thing is that you learn something, you get some answers to some pressing questions you’ve been thinking about for a long time. For me the main benefit is to be allowed to do this kind of work, to go out and check if this is really how things are.


    HS: Do you have a sense of what this paper gets cited for, mostly?

    AM: Well, I have to admit, this is not really something I am checking. Of course, I can go to the Web of Science for example, or on ResearchGate and look up who cited it. But that does not really interest me. There are so many more interesting things to do. Sorry about that.


    HS: Did this paper have a big impact on the future course of your research?

    AM:  Yes, I actually published the year after, which was a continuation of this study. And I was able to do that only because I was studying swallows and not widowbirds. As I said in the beginning, barn swallows are unique because all survivors return to the same site where they originally recruited. Therefore, I could go out in the subsequent year and check if the experimental manipulation had an effect on their survival. And indeed it had. There were survival costs of increased tail lengths. So this way I could show that sexual selection had life history consequences. This was yet another step towards solving some of those old puzzles that had bothered Darwin and other people for a long time.


    HS: Do you work on sexual selection on barn swallows even today?

    AM: Yes, together with my colleagues in Italy, I have just got a review paper accepted, where we looked at all the studies of sexual selection in barn swallows. It might not sound like a big review but, in fact, there are more than 100 papers on this subject from different parts of the world – from Japan and China to Europe and North America.  What this review shows is that there are many characters involved in sexual selection, and which character is important differs among populations in different parts of the world. This in turn suggests that characters involved in sexual selection in different populations might have had a role in speciation.


    HS: It’s been 18 years since this paper was published - would you say that the findings are still, more-or-less, true?

    AM: Yes, I would say that. There have been several other studies since then, using a similar experimental technique, both in barn swallows and in other species, and, by and large their conclusions are the same.  The results in Table 1 about effects of tail manipulation on extra-pair copulations have also been vindicated in several subsequent studies of paternity, in barn swallows and other species. So many of the relationships we established in this paper turned out to be very general.


    HS:  Today, do we have a better idea of why females choose long-tailed males in barn swallows?

    AM: Yes, I think so. Starting with Darwin and subsequently followed up by Fisher, there was this belief that attractive males would be very good fathers. It turns out they are lousy fathers. And that’s true of many other species too. So today it looks like resistance to parasites, or the ability to raise strong immune-responses against parasites, is an important mechanism in many different animal species. Some studies suggest that’s perhaps the case even in humans.


    HS: And that’s the case in barn swallows as well?

    AM: Yes. I have done a number of different experiments investigating the abundance of parasites and micro-organisms, like bacteria and fungi, and found clear relationships with tail length. And . And with experimental manipulation of tail length.


    HS: Another unknown you indicate in your paper is whether the females that mate with long-tailed females are also of better quality. Do we know more about that now?

    AM:  Yes, that is also generally the case. This was Ronald Fisher’s idea that attractive males would mate with females in better condition because females in better condition could start reproducing earlier. This way an attractive male would also be able to start reproducing earlier. That is clearly the case in barn swallows, as shown by several studies. In this first study, I made the treatment before the males had obtained a female. I went out early in spring and tried to catch birds every day so they didn’t have a chance to mate before I had caught them. That way I could measure how long it took for males with different treatments to obtain a female. But it is also possible to make the same kind of experimental manipulation after the pairs have become established. And the surprising thing is that if you change tail length when males have already obtained a female, they don’t give up on the female they have obtained. This could also account for Malte Andersson’s findings. We know that in his study the female birds were already on these territories. So, it’s possible that in his study the birds were already mated, but some females were reproducing and other females were just hanging around waiting for better conditions. In other words, his manipulation was more a manipulation of female parental investment.


    HS: Do we also know how natural variation in male tail length correlates with reproductive success in the barn swallow?

    AM: Yes, .


    HS: Is this from your own work?

    AM: Yes, , from the work of my former postdocs, and from the work of some of my colleagues. We are now a group of scientists working in Spain, in northern Africa, in Ukraine, in Italy, in Poland and other places. As I said before we just had a large review paper accepted in Biological Reviews, in which we summarise this.


    HS: Have you ever read this paper after it was published?

    AM: It’s a little bit embarrassing but I am not sure I have. It reminds me of what I said earlier about the similarity of scientists and artists.  It’s like writing a poem. You have put your heart into it and written it down, but then you don’t want to look at it again. You have moved on to other things, you are thinking about something else.

    Occasionally, I go back to what I have written when somebody sends me an email with a question about a particular study. Then I go back to check what was actually done.


    HS: If you compare this paper to papers you write today do you find differences?

    AM:  Sometimes, I am a little surprised about how well it was written, because in those days it was much harder. It was the dawn of email. Okay, we had the first versions of Microsoft Word, so it was starting to get easier. Today, in most countries of the world, scientists have all the papers at their fingertips, but in those days, I remember, it sometimes took several months for a paper to arrive at your desk, after you have ordered the paper in the library.


    HS: Has your writing style changed?

    LK: I didn’t have any training in scientific writing. Also, though I studied English in high school, it was for a total of six years. Not much. But now, even though I have been living in France for 21 years and speak French fluently, I mostly converse in English. My wife and I speak English to each other and we speak English to the children. So I think my language has improved over the years for these reasons.


    HS: What would you say to someone who is about to read this paper today?

    AM: Usually, what I encourage my students to do is to read carefully and be critical. When I say critical, I don’t mean that the students should read it and say this is rubbish, but that they should always think how they can improve this. Are there ways in which we can do this better? If that’s the case, the student hasn’t wasted his or her time reading the paper.


    HS: Do you think there are ways in which this experiment could be improved?

    AM: We have done these kinds of experiments, perhaps, 20 times now. And we have changed the design. One thing we have done for example is to paint the outermost tail feathers white, instead of cutting them. That way, the bird’s flight is not affected but yet, the tail looks shorter to the female. Another thing we have done is to paint over only the white spot in the outermost tail feathers. This white spot is a weak spot because the feather easily breaks there. In fact, ectoparasites, such as chewing lice, like to chew on this white spot. It’s intriguing why there is such a weakness in a character that is so important for mating success.


    HS: Have there been any advances in technology that might allow you to do these experiments better today?

    AM: Well, at a certain stage I was wondering if we should simply make some simple video recordings of birds and show the video screens to females. I am sure this will work, but the main problem is that if you catch birds that feed on insects and bring them indoors very quickly deteriorate in condition So that’s been the main reason why we haven’t done that yet.


    HS: Is this one of your favourite papers?

    AM: I think it was a nice study. I don’t know if it’s one of my favourites though. Often it seems like my favourite papers never get published. The papers that I like the most, reviewers don't like at all! I have a computer full of such manuscripts. There is a Finnish journal which I think is .  I have been wondering whether, perhaps, we should establish a Journal of Unpublished Results!









  • Post date: 1 year 8 months ago
    Citation for this post: BibTeX | RIS

    In 2002, , , , , , and published providing evidence for low host-specificity of tropical herbivorous arthropods. This finding was based on data from over 900 herbivores from 51 plant species in a rainforest in New Guinea. Based on this finding, Novotny and colleagues provided new global estimates of arthropod diversity in the range of 4 - 6 million species, which was substantially lower than . Fourteen years after its publication, I spoke to Vojtech Novotny about the making of these findings and their validity today.

    Questions sent by email on 19th June 2016; responses received by email on 17th July 2016.


    Hari Sridhar: What was the specific motivation to write this paper? Was the fieldwork done with this paper in mind or did the idea for the paper come later, after the work was started?

    Vojtech Novotny: The paper is called “Low host specificity...” and the study of host specificity of tropical herbivorous insects was the primary motivation for the underlying study - the grant proposal (for US NSF) and the field study. So there had always been a plan to write such a paper. It was only unclear what form it will take and for what journal. At the end we wrote two papers, and . The latter benefitted from a plant phylogeny for our studied species – something we were not sure we could get at the beginning of the study. Further, the Nature paper also revised the global estimates of insect diversity that had always been based on host specificity estimates. In the first version of the paper (submitted to Nature), we only said that our host specificity results will lead to new, greatly revised, estimates of global diversity, but did not actually perform the calculation. Both reviewers and the editor suggested that we expand this section and revise the global diversity estimates ourselves, which we finally did. So that part of the paper was not foreseen when we were doing the field research.


    HS: This paper has seven authors. How did this group come together?

    VN: The authors include the team that wrote the original research proposal (Novotny, Basset, Miller, Weiblen), plus two postdocs who joined the field work (Cizek, Drozd) and a collaborator recruited during the project to help with the plant phylogeny (Bremer). This group (except Bremer) was involved – entirely or partially – in many previous and subsequent papers on the ecology of New Guinean insects, and to this day represents a collaborative team. Its senior members have complementary roles and specializations – Novotny and Basset: insect ecology, Miller: insect taxonomy, Weiblen: plant ecology and phylogeny and Bremer: plant phylogeny. The Novotny – Basset – Miller – Weiblen collaboration has been going on for 20 years now and remains focused on plant-insect food webs in Papua New Guinea (PNG).

    My collaboration with Yves Basset goes back even further. Many years ago, I had applied for a University of Leeds position to study the insects on fig trees in Krakatoa. Though I didn’t get that position, in preparing for the job interview, I read a lot about figs and got very interested in the system. Sometime later I met Yves at an entomological congress. Yves was also interested in insects on fig trees and over a pint in a York pub our New Guinea collaboration started.


    HS:  The dataset used in this paper seems unique in its size and resolution. Can you tell me how this data was collected and over what period of time?  

    VN: The study has – in my view - three features that distinguished it from others at the time (and got it into Nature), viz. (i) the large scale of the data set, (ii) direct study of trophic interactions by feeding experiments, and (iii) plant phylogeny used in the measurement of host specificity. The data set was collected over seven years of continuous work in PNG rainforests, where we were sampling one set of (live) plants after another, using teams of trained paraecologists recruited from local villages. This approach was described in our papers and .

    Another important aspect of the study was sustained presence of researchers – first Basset, then me, then also Cizek and Drozd. In this way this study created a basis of a research team that later expanded and presently constitutes the .


    HS: Was all the insect curating and identifying done in field? Where was the plant phylogeny work done?

    VN: The insect sorting to morphospecies and databasing was done at our PNG base, and the final taxonomy overseas, mostly under Miller’s leadership.

    Plant phylogeny, in those days, was much more difficult than it is today. For our paper, it was partly done as a part of PhD studies by Weiblen (Moraceae) and partly by Bremer who specialized in another taxon we were interested in - Rubiaceae.


    HS:  Did this paper have an easy-ride through peer-review? Was Nature the first place you submitted this paper to? Did the paper change substantially from the first submitted draft?

    VN: Nature was (as you might expect) our first journal of choice – and also our first attempt to send something to Nature from our PNG research. The manuscript was well-received by reviewers and the editor, Dr. Rory Howlett, who wrote in his response While all three reviewers find your work of considerable interest, as do we, they have raised points to which we would like to see your response... In addition to minor comments and queries, two reviewers, supported by the editor, wanted us to come up with revised global insect diversity estimates of insects based on our new estimates of tropical host specificity. In the original draft we submitted to Nature we had concluded by saying: Further, estimates of global biodiversity at ≥30 million species (Erwin 1982) need to be revised as they are based on the assumption of narrow host specificity of rainforest herbivores, which was disputed also elsewhere (Basset et al. 1996a, Ødegaard et al. 2000). Based on the reviews we received, we calculated and included our own estimates of global insect diversity in the revised version we submitted.


    HS:  Do the main findings of this paper – “low host specificity of tropical herbivorous insects” and “4-6 million of insect species globally” - hold true even today?

    VN: Both findings still hold, by and large. There has been a subsequent discussion and even controversy as to how this low specificity compares to temperate zone patterns (i.e. what is the latitudinal trend in specificity), based on and subsequent work, but the patterns on specificity uncovered here remain accepted. Likewise, further analyses of global insect diversity, including , came close to our estimate.


    HS:  Did this paper create a buzz - within academia and outside - when it was published?

    VN: It did have fairly significant popular science exposure, including BBC, Los Angeles Times etc. It has, currently, 548 citations on Google Scholar – pretty much evenly distributed since the time of publication to the present, suggesting that it is seen as a solid and still relevant piece of work, rather than generating a wave of excitement that would later wane. Interestingly – but perhaps not surprisingly – all media attention, and also a part of professional attention, focused on diversity estimates, i.e. the part of the paper that was developed only in response to the review process.


    HS: How important has this paper been in your career? Did it have an influence on the course of your future research?

    VN: It was very important as it validated my choice to focus on long-term ecological research in Papua New Guinea. I had left for PNG originally for a 6-month stay in 1995, which was extended to one year, followed by another year after only a 3-week break back in Europe, and followed by a series of half-year stays. By late 2001, when the manuscript was submitted, I had spent 4.5 of the past 7 years in PNG. This was not exactly appreciated by my employer, the . This paper was the final proof to the institute that my PNG rainforest adventure was not a complete madness and could be productive scientifically.

    Interestingly, the paper was less influential on the direction of our future research – it only confirmed that we were on the right course, studying plant-insect trophic interactions in rainforest food webs (not that we had any doubts about it). However, it did represent a closure of a particular research direction where, every year, we would pick another set of rainforest tree species for one-year sampling of their herbivorous communities, to add to our growing data set. After this paper we re-focused on beta-diversity of herbivore communities across large geographic scales, which in due course led to .


    HS: Have you ever gone back and read this paper after it was published? When you read this paper now, what are the aspects about it that strike you first? What memories does it bring back?

    VN: I have not read it in its entirety for a long time, but did go back occasionally to check some particular statement or detail. It is striking that after almost 15 years I am not aware of any comparably detailed replicated data set from any other tropical rainforest site, which, incidentally, is typical for many other studies of community ecology patterns. Plus, we still do not really know how many species of insects there are on the planet. There is also a degree of nostalgia generated by this paper for my early days of much more solitary research in PNG (with no mobile phones and email only starting to reach the island).


    HS: If you compare this paper, to papers you write today, are there any differences, e.g. in writing style?

    VN: One unfortunate, but rather predictable, development in my career since the paper’s publication is that I am writing fewer papers as the first author. This is compensated by being able to shape more papers, mostly written by my students, as the last author, but that is not as much fun as writing a paper from scratch. The 2002 paper was my first stab at a direct, concise, sparse style of writing, which I have tried to develop and improve in subsequent papers. So, although it has its imperfections, it was a start in the right stylistic direction.


    HS:  Have you had the opportunity to go back to these field sites after the paper was published? What is the status of these plots today?

    VN: These are the rainforest plots closest to our research station where I am writing these lines now. I have been visiting them regularly for the past 20 years. These are village-based conservation areas and doing great. Some of my collaborators, recruited from these villagers and trained as paraecologists, have become co-authors on some of the subsequent papers, including .


    HS:  Do you know what this paper gets cited for, mostly?

    VN: I do take note of the citations (also because this is still my most-cited paper). I have not analysed closely the papers that cite the study, but I gather it is primarily the low host-specificity pattern that gets cited.


    HS:  What would you tell a student who is about to read this paper today? What should he or she take away from it?

    VN: Students today will see that the paper is obviously rather primitive by current standards when it comes to plant phylogeny, but remains of high standard when it comes to insect data. Interestingly, we used these insect data to make some rather daring extrapolations in order to estimate global insect numbers. Despite the fact that these numbers are of interest to many biologists and non-biologists, we have not progressed much with making these estimates more precise over the past 14 years.


    HS: Is this one of your favourite papers?

    VN: It is one of my (three) favourites as it was the first milestone in the PNG research, brought results of general interest to both the wider public and researchers, and remains relevant to this day.






  • Post date: 1 year 9 months ago
    Citation for this post: BibTeX | RIS

    In 1998, , , Judy Rankin-De Marona and published describing findings from an 18-year field experiment on habitat fragmentation effects on rainforest trees in Central Amazonia. This paper was “publication number 183” from, arguably, the most famous habitat fragmentation study in the world till date: the , or BDFFP as it is more popularly known. Eighteen years after its publication, I spoke to William Laurance about the making of this paper, how he got involved in the BDFFP, and what has happened to the plant communities in these manmade fragments, since then.

    Date of interview: 17th June 2016 (on Skype)


    Hari Sridhar: How did you get into the Biological Dynamics of Forest Fragments Project (BDFFP)?

    William Laurance: I studied rainforest fragmentation for my PhD in north Queensland, Australia, so I was, of course, aware of the BDFFP. It was a very famous project. Then I got to meet – Rob Bierregaard as we call him - at an ESA conference, I think, and we decided to edit .  This book came out in 1997 and ended up doing quite well.

    Then we had a symposium at Snowbird, Utah about forest fragmentation, and a number of people from BDFFP attended. One of them was , who was the field director at that time, and he told me about a position they were advertising. I had just then finished a three year postdoctoral stint at the in northern Queensland and was looking for my next position. This position that Claude told me about carried with it a dual appointment - with the fragmentation project and also with the . So it was an interesting opportunity.

    The big doubt in my mind at that time was that, although I was very interested in forest fragmentation and something of a generalist, I really wasn’t a botanist. I am still not a very good botanist. But fortunately for me, my wife - Susan Laurance - is a pretty good botanist. So we ended up pitching to them that she comes and manages the herbarium and I would be employed full time as a, so-called, senior scientist with the project.

    My job was basically to analyse this very large dataset, in which a large amount of money and time had been invested, but from which little had been published. Judy Rankin, who initially ran that thing, was very good at getting the whole network set up and getting a lot of plants identified, but wasn’t a strong publisher. So I was basically hired to publish as much of that work as possible.

    And of course it was a great opportunity. I am not the kind of person who can walk around in the forest and say this is this species and that is that species. I had basic training in botany and had taken a very good plant ecology course, when I was a grad student at Berkeley, from , a very eminent botanist. So I wasn’t completely ignorant. And I had done some very basic plant ecology as part of my PhD – habitat structure, indicator species etc. But my strength lay in analysing and interpreting data, and I knew my ecology well.

    I was also involved in fundraising and ended up getting quite a lot of funding through the LBA (Large-scale Biosphere Atmosphere experiment in Amazonia) programme for the project. I think I ended up getting about 8 million dollars from them, and also a couple of National Science Foundation (NSF) grants and quite a few foundation grants. And eventually my position, which started off as a temporary position, became a permanent position when I was permanently hired by the Smithsonian. So that’s how it started.


    HS: Which year was this?

    WL: My wife and I moved to Brazil in early 1996. We had been married three years at that point, and didn’t have children until we had been married at least 10 years. So those initial years when we were in Brazil, we didn’t have a family.


    HS: Were you hired specifically to do plant-related work?

    WL: Yes, I was hired specifically to work with this gigantic tree dataset. Subsequently, I ended up getting grant money to census lianas (woody vines) as well, and there’s been some interesting work coming out of that too. Luckily, as I said, my wife was a pretty good plant ecologist. That position really needed two people - one for data analysis and interpretation and another who could identify the plants.  They had actually collected, from every single tree, a sample of leaves or, if possible, leaves and flowers. So the project needed someone who would be comfortable working with that material. Of course, there were a number of new species that were discovered during this project.


    HS: What were the contributions of the other authors – Judy Rankin-De Marona and Leandro Ferreira - to this paper?

    WL: Well, it was really Judy Rankin who supervised all the initial field work. She was hired to set that up and had a large team of people working for her. I think she worked on the project for about 8 years, maybe even longer. We included Judy as a co-author on many of the papers that we produced because of how much she had invested in the work. She really did a terrific job. Judy was very strong on, what you would call, data quality control. I mean she was very tough and rigorous with her field people, and had extremely high standards of data collection, which of course was fantastic for us. We benefited greatly from all her efforts.

    Another key contributor was Leandro Ferreira, my research assistant. He did a lot of the database management for me. We had this big , and I would say to him - Leandro can you please extract this, this and this data for me and he would do it. He was also involved in managing some of the fieldwork - staying on top of what our field teams were doing and liaising with them. Especially in those first couple of years, when I was just learning Portugese -I speak it quite fluently now - Leandro was very helpful in translating what I was saying to the field guys. Later on he also did a PhD under me.


    HS: Was the data used in this paper collected, specifically, to address the paper’s objectives? 

    WL: Pretty much, yes. It was setup as an experimental study of habitat fragmentation. In fact, the original design was for an even larger experiment, with many more replicates and larger replicates. But because the ranchers didn’t do some of the things they promised they would, we ended up with a somewhat smaller experiment than what they had originally hoped for.

    But still, the big advantage of that project - and it really is a huge one - was that we knew what was there before it was fragmented. The trees were sampled before fragmentation happened, or at least as it was happening. And in tropical systems, where there is so much heterogeneity and patchiness and rarity, having that kind of information is incredibly valuable. That was one kind of control.  We also had plots in nearby intact forests, which we monitored over time, which were a second kind of control.


    HS: How did the BDFFP start?

    WL: It was the inspiration of , when he was working for . At that time, there was a very heated debate going on about - whether it was better to have a single large or several small reserves (SLOSS). This was a very rancorous scientific debate that generated, what some people called, more heat than light.

    Tom had earlier done his PhD in Brazil on Amazonian birds, and, around this time, he realised two things: first, Brazil was about to do some large-scale deforestation in the central Amazon, to examine the economic feasibility of large-scale cattle ranching. And two, that there wasn’t much data behind the SLOSS debate.

    Tom is very politically astute. He connected with a number of Brazilian researchers, the cattle ranchers - these were actually government sponsored ranches - and the bureaucracy that was promoting this, and basically convinced them to retain a series of forest fragments ranging from 1-100 hectares, when they created these ranches. The original plan was to also have a couple of 1000 hectare fragments but that didn’t happen. Tom is still very intimately involved in the project and has also raised a lot of money for it. It was a very high-profile study. Anybody who was working on forest fragmentation knew about it.

    The early days were tough. Just driving out to the site was a real adventure - really rough roads, and in the wet season they would be going through ponds with four feet of water. It was crazy. Wild West kind of conditions. And the camps were very primitive.

    By the time we got there things were much better established. Although, one of the things I really love about working there is that the camps are still very rustic; very basic. They are just aluminium roofs, completely open on the sides, and you sleep in hammocks. If there were mosquitoes, you put a mosquito net over the hammock.  We had jaguars walking through the camps in the middle of the night.

    It was a fantastic place. We had a fleet of four-wheeled vehicles that would run back and forth between camps, usually on Mondays, Wednesdays and Fridays. Now we have radios at the site, but back in those early days there was no communication.

    A friend of mine – – who is now at the University of Michigan, was once bitten by a Fer-De-Lance, which is a serious venomous snake. And the next vehicle was coming only three days later. One of the Mateiros – that’s Portugese for forest men – walked 20-30 km, I think, out to the road, then hitched a ride to the nearest town, and called for help. But till help arrived, Chris was just lying in a hammock with his foot swelling continuously. A funny story here - another mateiro, who had been bitten previously by a Fer-de-lance, offered to spit in Chris’s mouth, because he believed that would cure him. Chris says that he refused, but as his foot kept swelling and swelling and there was no sign of help, he was beginning to consider letting him do it!


    HS: How many fragments were created?

    WL: Eleven were created, initially. But then we lost two of them – a one-hectare and a 10-hectare- because their edges weren’t maintained, and over time they just sort of became reconnected to the continuous forest. So today there are nine fragments.


    HS: Why are all the fragments square-shaped?

    WL: I think it was just to rule out shape as a complicating factor. Otherwise we would have needed more replicates to separate shape effects from area effects. In those initial days, fragment area effects was the main interest. Also, square is the easiest geometric shape to create I guess, with a bulldozer when you are clearing land.


    HS: Was this one of the first papers looking at the effects of habitat fragmentation on tree communities?

    WL: Yes, for the Amazon it certainly would have been, and maybe even for the tropics. There were earlier studies in temperate forests. But there hasn’t really been anything comparable to it in scale, even today. This is one of the most floristically rich regions on the planet. We average almost 300 species per hectare of trees, which is among the highest you find anywhere in the world. It’s an enormous challenge to work in a tree community that rich because it makes it difficult to identify everything. I think we have a total of about 1300 species across all the plots. This also means lots of rare species, which makes data analysis quite challenging.


    HS: Was this one of the early publications from the BDFFP?

    WL: There were publications before this – a number of book chapters and maybe a handful of papers in good journals. I think what we were able to contribute was many more papers in high-profile journals. And for us it was fantastic because we had this great dataset, and of course people were really interested in the Amazon. So it was a really good opportunity for me in terms of my career. I was able to get a bunch of papers in the big journals – Science, Nature, PNAS, Ecology, Conservation Biology, etc. When you have a really great dataset it is much easier to publish in those top journals.


    HS: After you joined the BDFFP, how long did it take for you to publish this paper?

    WL: Well, we started publishing almost immediately. I arrived there in early 1996 and I think, within the first 3 months, we had our first manuscript submitted, which went to Conservation Biology. It was based on some data that Leandro Ferreira had been working on. So that was . There was, usually, about a 6-9 month lag between the review process and when it got published, especially in those days before online publishing. We also published - the first paper from that project to be published in Nature or Science - in 1997. That was the biomass collapse idea.


    HS: How long did this particular paper take?

    WL: Generally speaking, I tend to write pretty fast. The data analysis itself would have taken probably a couple of months and the writing of the paper would, I suppose, have taken a couple of weeks. But there was quite a lot of work in getting to that point – deciding what kinds of models to use to estimate tree mortality, deciding what models to use to estimate tree biomass turnover, thinking about what questions would be interesting, etc. Because I had studied fragmentation earlier, I was familiar with this stuff. But I had less knowledge about the plant ecology aspects, so that was a pretty steep learning curve.


    HS: Writing a paper in two weeks is quite remarkable!

    WL: Well, I write pretty fast. Typically, I publish 30-35 papers a year, including a few popular articles.  Once I start writing a paper, I really stay with it.  I only work on one paper at a time.


    HS: Did all the analysis and writing happen in Brazil?

    WL: Yes, we were based in the city of Manaus. Not too long after we got there, my wife decided to do a PhD, on the impacts of roads on Amazon forest understory birds. Whenever I could, I would go out with her to field; basically, be her field assistant.


    HS: Did you do most of the writing for this paper?

    WL: Yes, I did all the writing.


    HS: Did this paper have a smooth ride through peer-review?

    WL: Yes it did. I remember it was favourably reviewed and we didn’t have much difficulty getting it published. I remember , who is a well-known tropical biologist, was one of the reviewers and her comments were positive. I think it was relatively easily accepted.


    HS: Was Ecology the first place you submitted this?

    WL: Yes.


    HS: Did the paper get coverage in the popular press at that time?

    WL: I’m trying to remember; that’s a good question. I think it probably would have, because I have always done a lot of popular writing. I am a real believer in communicating to all kinds of audiences. Not just to scientists. I am trying to remember now..I think we published something in a popular Portugese science journal called . I have to go back to my CV to remember more.


    HS: Do these fragments still exist?

    WL: Yes, they do, though the project has been struggling financially. We didn’t know it at that time, but in hindsight, the time I was working there were the good old days! We had quite good funding from NASA’s LBA project, as well as NSF, the Mellon foundation and various other philanthropic organisations.

    NASA not only gave us a lot of money – something like 8 million dollars - but also did many things for us. They are engineers and they are very “can do”. So they would ask us what we needed to make our project better. Better roads? Done. Better bridges? Done. Radios in our field site? Done. More office space? Done. It was like boom boom boom boom and it was done. So we really got a lot more than the 8 million dollars. At the same time, they were demanding. If they called you for a meeting you had to go, if they made a request you had to comply right away.  But we were okay with that because we were really happy to have that funding and support.


    HS: It has been 18 years since this paper was published.  Do the patterns you describe in it hold true even today? What have you learnt about fragmentation effects on tree communities since then?

    WL:  I have actually made some notes here, because I can think of about 6-7 different things we have learnt over time. I will try and run through them in some kind of logical order.

    So what I am most struck by, when I read this paper now, . One of the things we have learnt is that, not only are many trees dying in the fragments, but a lot of the trees dying are big trees. That was a paper we published in Nature in 2000: . Now we are still not sure why this is the case. One possibility is desiccation. Tall trees have hydraulic challenges in getting water to the top. Some of these trees can be 50-55 metres tall, so they experience a considerable amount of hydraulic difficulties. And if there is a lot of desiccation stress - people refer to it as vapour pressure deficit - they can get little bubbles in their water carrying vessels. Embolisms. And then their water carrying vessels stop working and they can die of dehydration.

    The other possibility is that as these trees get bigger they grow proportionally thicker and stiffer. If you imagine one of these trees on the edge of the fragment caught in a windstorm, because it is tall, with a large crown and inflexible, it is more likely to get knocked over or snapped in half. So that is one thing we have learnt - the big trees are particularly vulnerable.

    Another thing we have learnt is that . A site that has got more than one edge nearby – like a corner site in a fragment – is more impacted by edge effects, than one that has only one edge nearby.

    Then we have learnt that . For instance, if you are comparing a fragment that’s surrounded by cattle pasture, with one that has some regrowth forest around it, then the latter will have lower tree mortality.

    We also learnt that there are two major types of secondary regrowth that occurs in this area. One type comes up when you just cut the forest down and is dominated by a plant called Cecropia. The other type comes up when you cut and burn a forest down and is dominated by a species called Vismea, which is more fire-adapted. The latter kind of forest tends to be more species poor and scrubbier. Now, it turns out that . You have bats and birds flying back and forth between the fragments and the regrowth and feeding and defecating seeds So if you are a fragment surrounded by Cecropia, you get Cecropia regeneration, and similarly with Vismea.


    HS: Was the matrix around every fragment the same at the time of this paper?

    WL: The matrix has been dynamic over time. When we first went there, it was mostly cattle pasture surrounding the fragments. But because the soil was so poor, cattle ranching did not turn out to be very viable. In addition, ranchers were occasionally losing cattle to jaguars and disease. So the ranches were real failures, economically, and were progressively abandoned. Ironically, after that happened, the forest fragmentation project has been maintaining the fragments, by clearing a 100 meter-wide strip around each fragment, at periodic intervals, so that we can continue to run the experiment.

    Going back to our learnings since this paper. Another thing we found was that .  Particularly in the first 100 metres from the edges, you see these very dramatic increases in pioneers - we published that in Ecology as well. . Lianas are natural enemies of trees. Trees that are infested with lianas typically tend to have lower levels of growth, higher rates of mortality and lower rates of fecundity. Also, lianas, by climbing from one tree to another, can tangle trees together. And then if one tree falls, it takes down the other one with it.

    Understanding changes to the overall species composition of these fragments has been a real challenge. . We showed that, even though the species richness of the fragments was not significantly lower than in intact forest, fragments were more variable and their dynamics were really different. Their compositions were changing, and changing most dramatically in fragments with high rates of tree mortality and tree turnover.

    Two groups of trees were particularly vulnerable: first, the understorey sub-canopy specialist trees, which are adapted to live in very stable forest interior conditions.  They spend their entire life history underneath the canopy. So, if you think about it - we know that in a rainforest there is only 1-2% of PAR – photoysnthetically active radiation - that gets down into the understorey.  Therefore, these trees have to be very specialised in terms of their physiology - they tend to grow very slowly, have very dense wood and have very specialised photosynthetic machinery that starts up quickly to make use of little flecks of fast moving sunlight that reach beneath the canopy.

    Another group that does not seem to do well are species that are obligate outbreeders. You see these a lot in the tropics. These plants as a group are vulnerable because they are often dependent on animals for pollination and even for dispersal. So, if these animal partners decline in number in fragments, they also impact the plants dependent on them.

    There are two other things we have learnt since then – one is that environmental synergisms are important.  This does not really come from this study, because these fragments are experimental and therefore protected – from fires, from logging, from hunting. So, in effect, they are a best-case scenario. But I have seen it in my work with scientists in other parts of the Amazon.

    For example, - who at that time was in Michigan State University - where we used 14 years of remote-sensing data of 700 different forest fragments and looked at the frequency of fire as a function of distance from the edge.  We saw an extremely strong increase in fire frequency within a couple of kilometres of the edge, and within 600 m of the edge it just went up hugely. The reason is, because of tree mortality, you get a lot of branches and litter and stuff accumulating on the ground. The ranchers burn their pastures every year to control weeds and produce a flush of green grass for the cattle.

    These fires, often times, when they hit the fragment they will just keep burning, especially in drought years when the fragment is dry. The initial fires will kill all the forbs, the vines and many of the small trees. The canopy thins, more plants die and more plant material accumulates on the ground. And the next time a fire happens, the intensity is even higher, because it has more fuel. So what you see happening over time is that these fragments implode.  When you look at them using remote-sensing, you can actually see the fragments shrinking year after year. We call them ”ghost forests” because you have these white bleached trees surrounding a reduced patch of rainforest that has been whittled down by recurring fires. We also know that in a lot of places, fragments are selectively logged and hunted. That’s another kind of synergism.

    Finally, the last thing we have learnt: They were changing, and changing in a concerted way. At that time I was really scratching my head over this.

    One of the things we managed to do was get hold of an independent dataset, which a PhD student, who had been affiliated with the BDFFP but who had worked independently, had collected in intact forest. It was only three plots, but we compared it to our own interior forest data and found striking similarities – the winners in our intact forest sites tended to be winners in his sites, and losers in our sites tended to be the losers in his sites.  We on which he was the second author.

    There are lots of hypotheses for what’s causing the changes in the intact forest, but the one that we favoured was that it was a consequence of rising CO2 levels. We know that whenever there is environmental change, all species don’t respond in the same way - you get some winners and some losers. The pattern in our winners and losers was consistent with this idea of CO2 fertilisation, although there are other factors, such as drought and rainfall, are also likely to be important. So, that was in 2004.

    More recently, . The lianas were sampled initially in 1997 and 1998. We went back in 2012, 14 years later, and resampled them and found out that lianas had been increasing in intact forests by about 1% per year. This is really interesting, and this is consistent with what some other people have discovered in other places. But we probably had the best dataset that’s been published so far, involving over 35,000 lianas.

    Then the very latest thing we have discovered is that what’s happening in the fragments is not just a consequence of fragmentation, but also a consequence of some large-scale global change driver.  . Because we are seeing changes in intact forest – change in liana numbers, change in forest dynamics - we think what is happening in the fragments is a kind of environmental 1-2 punch. They are being influenced both by local-scale phenomena, such as edge effects and area effects, and also by large-scale regional or global drivers, of which CO2 fertilisation is certainly the most prominent hypothesis.

    So we have learnt a lot, you know, since the time that 1998 paper was published.  What I am really struck by is how much more we know now. We certainly don’t know everything, we still have lots of questions to ask, but that paper really described some of the fundamental patterns, in terms of basic forest dynamics. Now we have realised now that there is a lot more going on. And we wouldn’t have discerned any of this if it weren’t for the large-scale and long-term nature of this project. Or if we didn’t have those controls in the intact forest.


    HS: Are these plots monitored even today?

    WL: Yes, pretty much. We tend to census the plots at about 5-year intervals. Each census is usually a 2-year job because it involves censusing all the trees in 69 one-hectare plots. And, of course, it’s an expensive undertaking, so we have to get funding for that. That’s something the project is increasingly struggling with. These long-term datasets are extremely valuable but they are also expensive beasts to maintain. I think a great deal of credit should go to Tom Lovejoy for all his efforts over the years to keep this thing funded.  I and others have certainly also contributed quite a lot of grant money to the project.


    HS: Were the papers that came after this one, based on updated datasets?

    WL: Very much. The re-censuses were incorporated with the prior data and we also did a lot of work since then on identifying material that had been collected but had not yet been identified. When Susan and I got there, the herbarium was in quite a state of disarray. A lot of material had actually rotted.  Also, at that time what they did was they collected three examples of the leaves from each individual tree, kept one in the herbarium and sent off one each to the New York Botanical Garden and the Kew Royal Botanical Garden.

    It turns out that in Kew, the boxes with the specimens were never even opened. The project had sent them there in the hope that their experts would work on them and identify them, but these boxes had simply been stuck away. So when we realised that a lot of material in our herbarium had rotted away, we paid to have all the material in Kew shipped back to us. And then everything was referenced back to a particular tree tag number, so that we could then go back and identify a lot of the things that had not yet been identified.

    My wife and her team made a lot of progress in terms of identifying those trees. Also, doing subsequent censuses of the plots as well as including lianas as a new group. We also subsequently have done a major survey of the small trees. We got an NSF grant to sample, in a subset of 20 of our plots which were carefully stratified across intact and fragmented sites, the small trees down to 1 cm.  we actually have those data all ready to analyse. I am just trying to find some time now to analyse that. Because I am sure there will be a whole series of papers that will come out of that.


    HS: Do you still lead the analyses on the plant dataset from this project?

    WL:  Well, yes, I am still the curator of this project’s plant data. We refer to this as the phyto-demographic project. There is a herbarium manager in Manaus who has a small team that works with her now. Then there is a field director in Manaus and there are still some field technicians though not as many as earlier. In fact, we have lost two-thirds of our staff because of funding cutbacks.

    We don’t, by any means, have a closed data policy. We collaborate with many people and many students and research groups use our datasets. But a data curator is important to make sure two people or groups are not doing the exact same thing. Particularly, if one of the two is a student. You have to kind of protect the student, because if someone comes along and publishes something that the student had spent 3-4 years working on, that is real damage to them.

    So one of the things I try to do when providing the data is to protect the interests of the students, so that they don’t get scooped by someone much more experienced, who could get the results out really quickly. The other thing we found was that it is, generally speaking, very useful to have one of our team members as collaborators. We found that when people tried to interpret the data without any input from us, it often ended in misinterpretations.

    There is a person, who I won’t name, who just had a paper that had earlier been rejected by two journals. I really had problems with the way he had interpreted the data. I told him what the problem was and kind of sat down and rewrote it. He then submitted it to another journal and it was accepted right away. So, as I said to him, there is a reason why it is often good to have some of our team members involved, because we can help with the interpretation, and we know the literature really well. I think part of the problem with this person was that he did not know the literature very well, and was in some sense trying to oversimplify the story. Once it became more nuanced, it was immediately accepted.

    Most authors are happy to collaborate with us because we give them a lot of help. Often times, the information they need takes quite a lot of effort to extract and convert to the format they want it in. Then there is a lot of information, on the history of the fragments and other kinds of stuff - supporting information we provide – that really strengthen the manuscripts.


     HS: These fragments are 18 years older than at the time of this paper. Do you think the patterns with forest age are different now?

    WL: Well, I think we are still seeing an increase in pioneers over time. And there is perhaps some evidence now that the rates of mortality might be starting to drop off a little bit. Originally, we thought that mortality was going to be chronically elevated. But I think what’s happened is that you have got a kind of selection process on the edges of the fragments - the most vulnerable trees have died, and the ones left standing are the hardy ones, which can tolerate drought and withstand the wind-shear forces that get whipped up around the edges of fragments. There’s also some effect of the matrix that complicates things, but I think we are starting to see some modest reduction of the tree mortality rate. Although, it is still considerably elevated relative to intact forest.


    HS: In the paper you say that edge effects are more important than area effects. Is that still the case?

    WL: Yes, I think that still holds, at least over the time scale that we have looked at it. We are looking only at the first 3-3.5 decades after fragmentation. Over the longer term maybe things will change.  Also, for other groups, like birds perhaps, fragment area might already be having a greater influence. And that can, indirectly, affect trees, in the case of bird species that are important seed dispersers. And area also interacts with edges because small fragments are proportionately more edgy.


    HS: Today, if you were to design this study again from scratch, would you do it differently?

    WL: Probably, yes. In hindsight, it might have been better to have had a larger number of plots but of slightly smaller size. The one-hectare has become kind of the standard in a lot of studies. That is good in the sense that it allows a high degree of comparability.

    But I think that it would have been useful, given the finite amount of money and effort, to perhaps, have had four times as many plots with each being 50*50 meters instead of 100*100 meters. It would have given us more replication. At the time they set this project up they didn’t anticipate how important edge effects would be.

    As it turns out, we don’t have a lot of plots in the 100–300 meter range from the edge. That would have been useful in teasing out some of the nuances of the patterns. But you know, hindsight is always twenty-twenty. No matter what, it’s a terrifically important project, it’s been reproductive and the value of these datasets just increases over time.  They become these long-term barometers of change and we are still learning lots of new stuff from them.


    HS: Do you think these fragments have done better or worse than what you anticipated 18 years ago?

    WL: I don’t quite know how to answer that. I think fragmentation has had a substantive and dramatic impact on the tree communities. But because they have been protected from other kinds of disturbances, it hasn’t been a complete catastrophe, like we see in other fragmented landscapes.

    The one thing I will emphasize is that, even though these fragments are different ecologically from intact forests, they are still very important to retain. Fragments perform some extremely important functions. They can be the last refuges for locally-endemic species. They can be stepping stones for wildlife in a landscape. There have been studies using genetic markers that have shown that tree pollen from some of these sites are from several kilometres away.  Clearly, there are bats and birds that are moving large distances across the landscape.

    Finally, fragments can serve as foci for recolonization of species in a deforested landscape. The bottom line is that fragmentation has very serious impacts but it is still important to retain all fragments. There’s no such thing as an unimportant piece of forest or an ecologically useless forest. It’s all valuable.



  • Post date: 1 year 9 months ago
    Citation for this post: BibTeX | RIS

    In 2005, , , , , , and published a paper in Nature describing their discovery of the unique reproductive system of the little fire ant (Wasmannia auropunctata): males and queens reproduce clonally, while the workers are produced by normal sexual reproduction.  While clonal reproduction in queens has been reported before, from other ant species, this is the first report of male clonal reproduction. When Fournier and colleagues genotyped the sperm in the spermatheca of queens of the little fire ant, he found  it to be identical to that of males. Based on this, they proposed that, in fertilised eggs that develop into males, the paternal genome has somehow eliminated the maternal genome after fertilisation. What is particularly interesting about this study is that it wasn't planned - the authors stumbled upon this discovery during the course of a conventional population genetics investigation.

    Eleven years after the paper was published, I spoke to Laurent Keller about the origins of this study, the story behind the discovery and the impact this paper has had on our understanding of reproductive systems in social insects.

    Hari Sridhar: You say “While conducting a genetic population study of this species, we discovered a new genetic system in which females and males both reproduce clonally”. What was your motivation to study this species in the first place?

    Laurent Keller: We were interested in the little fire ant because it has been introduced in many parts of the world. We wanted to conduct a population genetic study to see from where they were introduced and how they moved, to check if we could track from which population a new population originated. To do this we wanted to look at quite a few populations worldwide.


    HS: When did you first realise that there was something unusual in the genetics of this species?

    LK: We obtained very unusual microsatellite data for the first population we looked at. There were fixed genotypes which were not in . That’s when we realised that something was special.


    HS: Who did the genetic analysis?

    LK: Denis Fournier, first author of the paper, who was a post-doc. at that time.


    HS:  When you noticed this unusual pattern in the data, can you give us a sense of the discussions that happened among the authors? What did you make of it? What were the explanations you considered?

    LK: That was a bit strange – initially, we had some discussions, but we didn’t know what to do. The people I was working with, in the university in southern France – Fournier was working there at that time –, thought that maybe the pattern could be explained by one male reproducing with all the females in the population. Because I know the biology of ants I knew this was not possible, but I didn’t have an alternate explanation. So I took all the genotypic data and spent a few hours to think about what could be the possible mechanism. Then I went there, and we had a meeting, and I talked about my idea that males may be reproducing clonally. That’s when we decided to test this further, by genotyping the sperm in the spermatheca of queens and comparing it to the genotype of the males.


    HS: The paper has seven authors. Can you tell me how this group together?

    LK: Arnaud Estoup and I started the project. Arnaud wrote up the proposal and it was funded by a granting agency in France.  Arnaud was in charge of organising the collection of all the samples and the lab work and I mainly dealt with understanding the data that emerged from the study. Denis Fournier did most of the collection and all the lab work. He was a post-doc. co-supervised by Arnaud and me. The others on the paper, I am not really sure now. Some helped with the collection of samples.


    HS: In the paper you also use some data from New Caledonia – was that collected for a different project?

    LK: No, that was collected for the same project. We couldn’t get enough males and queens from the French Guiana site. That’s why we also included information from another site.


    HS: Where did all the lab work happen?

    LK: In Arnaud Estoup’s lab in France.


    HS: Did this paper have a smooth ride through peer-review?

    LK: It went through quite easily. The reviewers were all very positive. They made a few good suggestions which we followed and it went quite smoothly. I think it was only a single revision.


    HS: In the paper you say that this is the first report of clonal reproduction by males in ants, and probably only the second report in any animal. Have other cases been discovered since the paper was published?

    LK: Yes, there have been two more species - two other ant species -which have exactly the same mode of reproduction. One was discovered just after our study actually. That’s an interesting story. There was a Japanese scientist who already had data that was quite interesting – he had microsatellite data of queens, males and workers of an ant species called Vollenhovia emeyri, in which the males never had alleles of the mother. He sent me a manuscript in which he had suggested this pattern arose because every time males were produced there were mutations at those microsatellites.  I felt that that was not very likely, and told him about our own similar findings and my explanation for it. I asked him to send me the data, and after looking at it, I told him that he most likely had the same type of social organisation that we discovered. I sent him a copy of our manuscript and asked him to do a few more analyses to confirm what he had, which he did .

    The other case was also interesting. I saw a poster of a master’s student reporting data on 12 microsatellites for the longhorn crazy ant (Paratrechina longicornis).  She told me her Master’s project had failed because none of her microsatellites worked - she got very different genotypes for males, workers and queens. I told her that maybe it’s interesting and asked her to send me the data. After looking at it, I told her that her species most likely had the same type of reproduction we had just uncovered. I discussed with her and her boss and proposed that they continue the study. However, the student finally decided not to continue in science and so we hired a post-doc. – Morgan Pearcy – to continue the work. He collected more data, confirmed that it was the same mode of reproduction as ours, and .


    HS: Do you think this finding opened up a new line of research - a new way of thinking about reproduction in social insects?

    LK: Yes, I think people are much more open-minded now, to the idea that there may be a lot more diversity in reproductive systems in social insects, than was thought before. My guess is that maybe as many as 10-20% of ants, and probable many other social insects, show unusual mode of reproduction.


    HS: Has there been more research on the little fire ant after this paper?

    LK: Yes, the group from France has done more, to see if they can find a population where there is sexual reproduction; . They also conducted several other studies on the biology of this ant later on.


    HS: But has the main finding from this study – clonal reproduction of males and queens –been contradicted by later studies?

    LK: No. In fact, it has been confirmed by later work.


    HS: In the 11 years since this paper was published, have you ever read the paper again?

    LK: Not really, except, occasionally, when we need to cite it somewhere else. Then, I check it to see exactly what we said.


    HS: Did the paper create a buzz when it was published?

    LK: There was some stuff but I didn’t follow it too much. There was a , and many other journals mentioned it – Current Biology, Nature Genetics maybe.  I don’t remember exactly which one, but there were quite a few which mentioned our study.


    HS: When you did the work, did you anticipate that it would attract a lot of attention?

    LK: Yes, we did think that people will be interested by this.


    HS: Did this paper have an impact on your future research?

    LK: It’s difficult to say, I was already interested in unusual modes of reproduction at this time, and so it, probably, just strengthened my interest in this type of research.


    HS: What would you say to a student or researcher who is about to read this paper today? What should he or she take away from it?

    LK: First, of course, the data itself. But also that, when you do a study, . As I mentioned earlier, the Japanese scientist also had similar data to ours, but because he never thought that a son could have no genes from the mother, he had to come up with some strange and unlikely mechanism to account for the data. So, I think what’s really important is to be open-minded enough to expect strange results. Then, to think carefully about what it could mean. And finally to take into account not only what’s known, but also what’s unknown.


  • Post date: 1 year 9 months ago
    Citation for this post: BibTeX | RIS

    In 2001, Nicholas Gotelli and Robert Colwell published that flagged common mistakes that ecologists were making in measuring and comparing species richness, and suggested ways to avoid such mistakes. Even today, 15 years after its publication, Gotelli and Colwell 2001 remains a must-read for anyone interested in richness and diversity research. I interviewed Nick Gotelli on the making of this paper, and the impact it has had, on his research career, and the field of species diversity studies.

    Date of interview: 16th June 2016 (on Skype)

    Citation: Gotelli, N. J., & Colwell, R. K. (2001). Quantifying biodiversity: procedures and pitfalls in the measurement and comparison of species richness. Ecology letters, 4(4), 379-391.


    Hari Sridhar: I’d like to start by asking you what your motivation was to write this paper.

    Nicholas Gotelli: There were actually two motivations. In a way, I had written an abbreviated version of this paper in a chapter on rarefaction in my 1996 book with Gary Graves . That chapter laid out some of the ideas from the literature that were already there. The second thing happened a little closer to the date of the paper. I was working with my graduate student (now a Biology Professor at St. Michael’s College), analysing an experimental set of data on the effects of disturbance frequency, intensity, and area on the biodiversity of stream invertebrates. We were interested in aspects of the (IDH) - trying to get at the question of whether disturbance increases or decreases species richness. It turns out that the answer depends entirely on whether you use a rarefaction approach or not. Basically, you can get opposite answers depending on the exact way that you measure species richness. So, I think it was these two things that got me started on the review. Those analyses appear in .

    I had also started collaborating and talking again to my old mentor . This paper is the first one we wrote together, but Rob was my undergraduate ecology instructor at the University of California, Berkeley, in 1979, where he taught an upper division course in Community Ecology, which had a huge influence on my career path. So, a few years later, when I began working on , I started corresponding with Rob again, because my lecture notes from his course (which I still have) were critical for the chapter on interspecific competition. That book was what brought Rob and me back into contact, and then I began working on the rarefaction problem for the null models book and again continued talking to Rob. Of course by this time, Rob had published , which are also an important part of this story.


    HS: There are two ways in which Ecology Letters gets submissions for its Review articles section: unsolicited and through invitations. Which was it in the case of this paper?

    NG: This is a long time ago and I’m actually a little vague on the history here. was in charge of Ecology Letters at that time and, to be honest, I cannot remember whether he solicited us or we went to him with this idea. We may have gone to him, I am not sure. I think we may have set up the agreement to do it, but the pieces were already there. One of the interesting things about this paper is there is very little in it that was new, even at the time it was published. In that sense, I think it is kind of an odd paper, because it’s really talking about ideas from the literature that were 20 or 30 years old.


    HS: Can you tell us a little about the writing process itself – did you get together with Rob to write this, or was it done mostly over email?

    NG: No, as with almost all my papers with Rob, everything was done over email. And in the late 1990s, there was no “track changes” feature in MS Word, but there was text highlighting. We would each use two highlight colours – one highlight colour was for your new text and one highlight colour was for your comments on the existing text. When you received a copy from the other author you went through, and for the new text, if you liked what they wrote, you “decolorized” it, as we used to say. I remember manuscript drafts would get huge in size with the 4 rainbow colours of highlighting, but gradually it would shrink down as we finalized the wording.

    Rob is a wonderful collaborator and correspondent, and I think our comments, in some cases, ended up being longer than the actual text. He really is a fabulous writer. He takes great care in the craft of writing, and a lot of the time was spent reworking particular sentences and organizing the different sections. That’s how it always is writing with Rob. In a way, it was not a hard paper to write at all. As I said, most of the things were already in the literature in one place or another, but we both felt that ecologists were not paying attention to this older literature. They were not taking into account some of these really basic issues on how to standardize and quantify species richness.


    HS: Did this feeling come from seeing a lot of papers using richness measures inappropriately at that time? 

    NG: Yes, there were lots of papers that were, you know, calculating diversity indices, species per unit area etc. - these types of calculations. Some of this was in the conservation literature, where people were trying to extrapolate species-area curves to estimate extinction rates and things like that. But making these linear re-scalings always causes problems for species diversity measures.


    HS: How long did it take – from when you had the idea to write this paper, to its publication?

    NG: It wasn’t that long. I am going to put it at somewhere between six to nine months.


    HS: Did it sail through review?

    NG: Yes, I remember that the reviews were pretty positive. We had a few suggestions from reviewers for improvement, but there was no controversy surrounding the paper at all. The reviewers seemed to appreciate the thorough look at the history of this topic, as well as the fact that the paper had some fairly specific advice on how people should go about trying to analyse species richness and species diversity.


    HS: In the Acknowledgements, you mention the grants you and Rob had obtained to develop and respectively. Was the development of these software happening around the same time?

    NG: Yes it was. Rob had been developing his EstimateS software, and I was in the early phases of developing the original EcoSim with Gary Entsminger. At that time, those programmes were fairly novel. There wasn’t as much software available then as there is today. This was well before the days of R and the open-source revolution, so there certainly weren’t that many pieces of free software available. We were excited about that aspect of it as well - to have these computing tools that we could introduce at the same time when we talked about the theory.


    HS: Am I right in saying that this is one of the most-cited papers in ecology, at least in the recent past? Did you anticipate the paper to attract so much attention and so many citations?

    NG: Yes, the paper has been very well-cited. But we did not anticipate it would become so popular because, as I said, from our perspective this was only a review article. What’s more interesting than the total citation count (2666 citations as of 4 July 2016) is the profile of citations through time. From 2001 through 2014, each year this paper received more citations than the previous year, and it is still getting 100 to 200 new citations every year. That’s a very long half-life.  Most things in the literature now are so short-lived that you wonder what their impact is going to be, and if they are actually going to be cited in a few years. That this paper continues to be well-cited after 15 years is very satisfying for Rob and me.


    HS: Do you have a sense of what your paper gets cited for, mostly?

    NG: The Ecology Letters review seems to get cited in 3 kinds if papers. First, papers where people are developing new theory and models for estimating species richness. Second, it is cited in plenty of empirical papers that are specialised on sampling of particular taxa. And third, it’s cited in papers that are talking about the fact that diversity indices are so sensitive to the amount of sampling that takes place.


    HS: In the 15 years since this paper was published, have you ever had the need to go back to this paper? Or is this the first time?

    NG: I’ve occasionally gone back to the discussion of individual-based and sample-based rarefaction. The paper introduced a useful taxonomy and description of how to talk about these sampling designs. One of the novel things introduced in the paper is the idea of adjusting or shifting the sample-based rarefaction curves to put them back onto a scale of abundance. I think it’s very useful for teasing apart the effects of abundance per say versus the shape of the species-abundance distribution. And that’s a good way to work with sample-based data.


    HS: When you read this paper now, what strikes you about it? Is the writing style different from the papers you write today?

    NG: No, I don’t think my writing is very different from today. This is how I usually end up writing, especially when I’m working with Rob. He brings out great clarity in the writings of his co-authors. He never takes anything for granted and he doesn’t want anything to hide behind jargon. So when you are writing with him you have got to explain yourself clearly and thoroughly. He won’t let anything get by that’s not crystal clear in the writing. Those of us who collaborate with him are very lucky to have had that influence on our own writing.


    HS: What kind of impact do you think your paper has had? Do you think there has been a reduction in inappropriate usage of richness measures after your paper was published?

    NG: Yes, in terms of reducing mistakes, I think it has had a positive impact. Reviewers, today, routinely expect authors to standardize or rarefy their data when comparing biodiversity. So I think that message has gotten out. One of the great things about working with Rob on this paper was that it led to a long-term collaboration with Rob and .  That resulted in some very important theoretical extensions and new forms of analysis based on the rarefaction idea. Since the 2000s, there has been a kind of a renaissance in the literature on how to estimate and measure species diversity, and there are still lots of new and interesting papers coming out on that topic. I guess I would say to people who are reading our Ecology Letters review for the first time - recognize and pay attention to this whole new literature that came after this paper!


    HS: One of the things you emphasize in your paper, towards the end, is the need for more research on asymptotic estimators. Has that happened?

    NG: Yes. The estimator that is still the most popular and widely used is the set of Chao estimators. They are popular for a couple of reasons.  First, they are easy to calculate. Second, they have very good statistical grounding, interestingly based on the computer science theorems from ’s work during World War 2. Turing cracked the Enigma– the German coding machine – by developing theorems and formulas to estimate the frequency of undetected categories. And that’s exactly the basis for Anne Chao’s estimators. These estimators have performed pretty well in comparison with alternatives such as jack-knife estimators. However, there is still room for improvement in rarefaction and asymptotic estimators.

    One thing people don’t like about rarefaction is that, in a way, it forces you to throw away all your data that is above your worst sampling level, in order to make everything comparable. And no one wants to throw data away. So in 2012, in the Journal of Plant Ecology, Rob and Anne spearheaded . In that paper, we showed that the rarefaction curve, which is the interpolated part, can be smoothly extended and linked up to the asymptotic estimator, which is the extrapolated part. So instead of having to worry about which particular part of the curve you are going to compare, you can actually visualise the entire curve and the uncertainty that’s associated with it. I thought that was pretty powerful.

    , and also, has gotten at the idea of coverage-based rarefaction and sampling: it’s not just the number of individuals that matters, but where you are on the rising part of the rarefaction curve. And your adjustments can be made that way.

    The asymptotic estimators - the Chao estimators - are great, but, as Anne Chao has repeatedly written, these are only minimum asymptotic estimators. Some people have been unsatisfied with them  when applied to “big data”, such as genomic datasets or surveys of hyper-diverse microbiomes. There is research ongoing to develop better estimators to use with hyper-diverse faunas. These are just some of the extensions that have followed from the framework laid out in the 2001 Ecology Letters paper.


    HS: Does the statistics of diversity and species richness continue to be one of your primary research interests?

    NG: Yes. Thanks to this first paper with Rob, I have been collaborating with Anne Chao and other researchers on a number of projects related to biodiversity estimation. For example, in , we dissect the effects on rarefaction curves of changes in species richness vs. changes in species composition. I am also in a working group led by at , where we are developing a set of steps for comparing diversity curves in replicated experiments or diversity surveys. So biodiversity statistics is still a very active research area for me.


    HS: Was this the first review paper you wrote?

    NG: It may have been the first review paper I wrote, but it came just after the null models book, which ended up being a huge review on all sorts of different topics in community ecology. The literature at that time was not as vast as it is now, so by the time and I finished the null models book, I felt I had a handle on the community ecology literature as a whole. Unfortunately, I don’t feel that way anymore.


    HS: Do you enjoy writing review papers?

    NG: Yes. They are a chance to pull things together and organise them and make a case for how things ought to be done. I think the goal of a review paper is not simply to exhaustively step through everything that’s in the literature, but to organise it, simplify it, and pull it together. Basically, to provide a framework for it. It’s not just this collection of papers that are cited one after another in a linear sequence. There is a certain development to the ideas that are contained in them, and I enjoy organizing that material into an explicit framework.


    HS: Was the material presented in this paper also published in other forms later on – e.g. as a book chapter?

    NG: Not that I am aware of. But there’s so much secondary publication that goes on these days, it could be that it has happened and I don’t know! In 2014, Rob and I contributed a chapter to the excellent book , in which we provide an overview covering some of the same topics on the estimation of species richness. That’s a little bit more up-to-date, as it includes recent improvements of the estimators and the theory that Anne Chao has developed since the Ecology Letters review


    HS: Is this among your favourite papers?

    NG: Yes, but I feel a little twinge of guilt about its popularity because we were mostly reviewing ideas that were 20 or 30 years old. I’m grateful that it has been cited well, and I’m pleased because I think it has had a positive effect on how people analyse biodiversity data. And, of course, all my papers with Rob are near the top of my list of favourites. But this one seems odd to me because of its relatively small content of truly new information.


    HS: Thanks Nick. That covers all that I wanted to ask you. Before we end, I have to tell you that all the three textbooks you have written - A Primer of Ecology, and Null Models in Ecology – have had a big influence on my research. I keep going back to them even now. Thanks for that too.

    NG: That’s very flattering, thank you. You know, it’s funny. I was working on those books as an untenured assistant professor, and I had some senior faculty members ask me whether I knew what I was doing -  was it a smart thing to be writing books at this stage in my career? In retrospect, it was a very smart thing to do. Unfortunately, I’m probably better known for those books than for my research papers!

    The statistics book grew out of a graduate course I taught without any computers. It was all done on the chalkboard. Many of the ideas for that book actually came from my notes for that course. For that book, I joined forces with another wonderful collaborator, Aaron Ellison, whom I have worked with for the past 20 years on the ecology of carnivorous plants and temperate ants. The stats primer was a lot of fun to write with Aaron, and we completed the first drafts very quickly.

    In 2012, Aaron and I published a second edition to the book in which we added two chapters, one on mark-recapture data and one on the estimation of biodiversity. And in 2008, I published a 4th edition of the Primer of Ecology that added a new chapter on biodiversity estimation. So, I am happy that all 3 of those books that you have mentioned have key chapters that are built on the framework developed in the 2001 Ecology Letters paper.






Powered by Drupal | Theme modified by Naupaka Zimmerman from Danland by Danetsoft | | INNGE is supported through a collaboration with INTECOL