In 1984, Gerald Wilkinson published a paper in Nature showing that vampire bats share food in the form of regurgitated blood, within groups that contain both kin and non-kin. This was one of the first clear documentations of reciprocity in animals, and the paper went on to become a citation classic and a textbook example for reciprocal altruism. I spoke to Gerald Wilkinson about the making of this paper, and its impact and relevance in the relation to whats happened in the 32 years since it was published.
Citation: Wilkinson, G.S., 1984. Reciprocal food sharing in the vampire bat. Nature, 308(5955), pp.181-184.
Cited by 858 (source: Google Scholar on 1st July 2016)
Date of interview: 15th June, 2016 (on Skype)
Hari Sridhar: Did you go to Costa Rica with the intention of studying bats?
Gerald Wilkinson: My initial trip to Costa Rica was as a student, as part of an Organisation for Tropical Studies field course. But at the same time, my advisor and his wife were also in Costa Rica, doing research on bats, and so I visited their site before the course for a couple of weeks. After the course, I got money from the OTS to stay another three months and continue a project that I dreamed up at the site where they were working. This was not on vampire bats but on a different kind of bat. I had this idea about fruit dispersal and group foraging in a little bat that both feeds on nectar and fruit. But these bats were very hard to study because they were very easily disturbed and would fly out of the roost whenever I tried to catch them. Vampire bats, on the other hand, were really common and very easy to work on. Around the same period of time - those three months - Jack Bradbury went to a regular bat meeting where he heard a German biologist - Uwe Schmidt - report that he saw vampire bats regurgitate blood to each other in his captive colony. Jack wrote me a letter and told me about that observation and said that, maybe, this might be something I wanted to think about. And so I actually began, at that time, trying to catch and band vampire bats, and then dreamed up the project that I subsequently did between '78 and ’83.
HS: In the paper you say a “26-month (September 1978– February 1983) study”. Does the 26 months refer to the time you actually spent in Costa Rica?
GS: Yes. There was the initial 3 month period after which I came back and wrote proposals to raise money for my work. So it was awhile before I was able to go back. On the next trip, I stayed continuously for 9 months, came back to campus for 3 months and then went back and stayed for a full year. Finally, there was a fourth trip that was only six weeks. It was on that very last trip when I did the reciprocal feeding experiment with the captive bats.
HS: Were you already registered for a PhD when you went to Costa Rica for the first time?
GW: Yes, it is pretty common in the US that students will first pick an advisor and then develop a research project after they start. And it varies, depending on the advisor, how independent these projects are - sometimes they are very close to what the advisor does, sometimes they are quite distant. We got money from the National Science Foundation (NSF) to support the project I did. Actually, money was given for two different projects – the vampire bats in Costa Rica and also for greater spear-nosed bats in Trinidad, which my advisor was going to do.
HS: You also mention another grant – the NIH training grant – in your paper. What was that for?
GW: That was a consequence of going to graduate school at University of California, San Diego (UCSD). They had - they may still have - one of the longest running training grants from the NIH, largely for genetics. So I was supposedly doing genetics! I was, in a sense - taking blood samples, doing allozyme electrophoresis, genotyping the animals, but I am sure I was the only person supported on that training grant who was doing field work. Everyone else was doing lab studies, of viruses and Drosophila and other more genetically tractable organisms
HS: Did you start a PhD immediately after you finished your Bachelor’s? How old were you then?
GW: Yes, right after my Bachelor’s. I was 22 when I started my graduate work, 23 when I began doing that vampire bat work and 28 when I finished.
HS: I’m intrigued by one aspect of your field work - “focal animal sampling within roosts”. These roosts were in tree hollows. How did you manage to observe these bats inside these hollows?
GW: The trees were large. Most of them were Anacardium excelsum - a relative of cashew - and they tend to get very large hollows. Most of the hollows were big enough that I could go into the tree. One roost was in a silk cotton (Ceiba pentandra) tree that was so big that I could lie down completely inside the tree.
HS: Inside the hollow?
GW: Inside the hollow, yes. The diameter of the tree was about six feet inside, probably about 10 feet outside. So these were big trees. The bats would be at the top of the hollow, which, in most cases, was pretty high - probably twenty to thirty feet above where I was. So, I would actually lie on my back and observe them with binoculars. I tried to video record them with Super 8 mm film but that didn’t work very well. With a camera you are stuck in one location, but the bats move around a lot so I needed to move around too to keep track of them. We also had to wear respirators when making the observations. Histoplasmosis is a fungus that grows in bat guano, something I was very aware of and concerned about. So the students that helped me and I always wore these respirators. It was not easy. It was very tiring so I always had someone helping me. One person would call out observations and the other person would sit outside the tree and write them down.
We also made little devices that would emit a click at set intervals of time, which we would listen for with earphones. I think for all the vampire work, I had set it to go off every 10 seconds. At every click we recorded whatever behaviour the bat was doing. I got the idea for this device from a primate field course I took as an undergraduate student at UC Davis. Davis is one of the eight places in the US that has a primate centre, where different species of monkeys or apes are kept in large outdoor enclosures. During that course we observed bonnet macaques in one such enclosure and recorded behaviours with the help of these clicking devices.
You know, at the beginning I was not sure exactly what I was going to find. I just knew I needed to be systematic, be quantitative. I knew I wanted to study food sharing but I didn’t know how often I would see it, so it was done with some level of hope. The information from these focal animal samples was mainly used in a . The food-sharing was so infrequent that I recorded it whenever I saw it. It was so rare that if I had done it only on focal animals I would have had no data. In fact, in the first six months, I think I had seen it only a couple of times. At that point I was starting to think that the whole project was doomed. I think part of the issue was that – and this is not mentioned in any paper –in order to see the animals and identify them, I used coloured reflecting bands – bird bands, basically - on the wings of the bats. And to see the bands inside the tree I had to shine lights on them. Initially, the bats would always hide from the lights and so I couldn’t make any observations. In order to overcome that, I would, every single day, take a miner light into the tree and shine it on the bats continuously. These would last for 12 hours with rechargeable batteries, and the bats had nowhere to go, so in some time they got habituated to the lights. I could confirm this with a night vision scope which I could use with infra-red light. Infra-red light is invisible to bats. So when I compared their behaviour with the night vision scope and with the lights I couldn’t tell any difference. But it took months.
HS: What were the bats feeding on?
GW: The site where I worked was a cattle ranch, but the bats, actually, never fed on the cows. They fed, primarily, on the blood of horses. And I know that because we put radio-transmitters on the bats and tracked them out to the pastures. Vampire bats do feed on cows in other places but this particular population seemed to like horses more than cows.
HS: You have acknowledged numerous people in the paper. Can you tell us a little about their roles?
GW: Yes, I had several student assistants. Terry Lamp, Robin Weiss - it was kind of funny because we were often called Batman and Robin - and Michael Lee Jones. During the two long stints of field work, one or two of those three people were always with me. And then in the final six week period a guy named Doug Bolger helped me. I think, all of them had just finished their undergraduate degrees, so it was kind of a break for them, before getting a job or going to graduate school. Actually, I’m not sure now, but maybe Michael had a little more experience. Maybe he even had a Master’s degree when he came to work with me.
Anyone else who I have acknowledged read and commented on drafts of the paper. This included Mark Taper, who was my grad. student office mate, Robert Gibson who was a postdoc at that time in the lab, and my advisor Jack Bradbury.
HS: If you don’t mind my asking – how come your advisor was not an author on the paper?
GW: No, I don’t mind at all. Jack has an unusual attitude towards his graduate students, in that he would not insist on being a co-author on their papers. He would agree to be a co-author only if he thought that he had contributed an equal amount to the work. In the case of this paper, he contributed to the writing, but I did all the work. In fact, those first two weeks before my OTS course was the only time we were ever in the field together. So he had very little idea of what I was doing until after I had done it. But I did ask him to be a co-author, because in my mind, I would have never done what I did if it weren’t for his advice. But he said - you did it, you deserve it, you should take all the credit.
HS: In your paper you also mention a personal communication from T. Fleming – “one pair of females marked by T. Fleming (personal communication) in 1970 roosted in the same area in 1981". Are you referring to ?
GW: Yes, that’s Ted Fleming. Ted had previously worked at the exact same site that I was at, and he had banded vampire bats, so some of the bats that I caught were ones he had also caught and banded over 10 years earlier. So I used some of that information to age some of the animals that I had.
HS: While you were doing this work, did you already get the sense that this was something big and important?
GW: I was highly motivated to to look for reciprocal altruism. That was definitely in my mind from the beginning. But I did not know at all, until almost the very end, that I had found much of anything. Even after we did the final experiment we did not know what we had found, because we didn’t analyse the data until we got back to San Diego. I was just randomly taking bats out and watching how they fed, and we were doing it all at night, and also doing other things at the same time, like radio tracking. So we were fully occupied in just doing field work and collecting the data. It was only afterwards that we said - Oh my goodness! Look at this! And it was only at that point that I thought – Wow! This is actually pretty cool.
I first sent that paper to Science but they said they didn’t want to review it. So then I sent it to Nature, who reviewed it but returned it and said they won’t accept it but would be interested in seeing another version – you know, that kind of standard letter. Now that was one of the first papers I had submitted so I took that as a rejection. But people explained to me that it wasn’t being rejected outright and that there still was a chance.
HS: At the time when you did this work, was the idea of reciprocal altruism topical in academia?
GW: Yes definitely. came out in '71. I learned about Trivers’s work as an undergraduate at Davis. The guy I took the primate class from ran a graduate course on E.O. Wilson’s Sociobiology. That would have been 1975, I think. I learned about reciprocal altruism at that time and became interested. And then, around the time I started my work was when , which was one of the first claims of finding some kind of reciprocal altruism in the wild. That got a lot of interest initially but also many people were questioning it. So, I was really very motivated to work on cooperation and look for some kind of reciprocal system but I didn’t know the vampire bats would do that. At that time I sort of assumed it was all going to be kinship, but I was very keen to test the two.
HS: I notice that you didn’t cite Trivers 1971 in your paper..
GW: That’s interesting, I didn’t remember that. Nature only allowed a limited number of references. Yes, I think it was 20 max. But I can’t imagine I didn’t cite Trivers! Well, part of it was - this was much later - the came out and I think I cited that instead. But Trivers should have been cited, because the motivation for this study was prior to Axelrod and Hamilton. I mean I was already down there watching the vampire bats long before that paper came out. But once it came out it actually gave me the idea for using the censusing as a way to measure the opportunity for the bats to share food with each other. That came directly from Axelrod and Hamilton. They have this little variable called ‘w’ which is sort of the likelihood that a particular pair will find the same circumstance in the future. Sort of the opportunity for future reciprocation. Earlier, I had some other method I had come up with to contrast kin selection and reciprocity, but when I read their paper it was like – Aha! I could just use my census data to calculate association. It was fortuitous because I did not have a specific plan for the census data at the beginning, except that it seemed like a good way to quantify social organization.
HS: Did the paper create a buzz when it was published?
GW: Yes, I think so. Certainly, among all the people interested in reciprocal altruism. But it did not get a broader reach for a while, apart from some coverage in newspapers - the LA Times and the Boston globe, maybe the New York Times. But there were several subsequent things that happened that increased its visibility. I got invited to write a , which reaches a much wider audience, including schools and non-scientists. That piece then, I think, motivated a . Right after the Nature paper came out, Trivers wrote a book called – this was in 1985 - and he called me up and interviewed me in great detail about the study. And to me, his description of the study is still the best of any that I did not write. He obviously knows the topic and he seemed to really like the study. Even though I didn’t cite his paper! I am sure he must have asked me about it at the time. His must have been the first book to carry my work, but after that, for a while, it was put in essentially every animal behaviour book and a fair number of general biology books. For a period of time it became the feature example for reciprocity, or something other than kin selection, which is funny because it does involve kin! I think it is partly also because it is vampire bats – they are sort of mysterious and sensational. So it sort of snowballed after that. It’ll be interesting to go back and track its citations over time.
HS: Were your findings and interpretations controversial when they were published?
GW: No, I wouldn’t say they were controversial then. I can remember interviewing for a job at Cornell and having a long conversation with , who was sort of very pro kin selection. I had conversations with Bill Hamilton. After my PhD I went to Sussex, England for a post-doc. and during that time I went up to Oxford and gave a seminar on vampire bat food sharing and Bill Hamilton and Richard Dawkins were in the audience. Actually that’s the other thing I forgot to mention, I should have remembered this - probably the one thing that catapulted my study to fame more than anything else was, after listening to me at Oxford, Dawkins immediately realised that this was a good story, and included it as an example for tit-for-tat in the second edition of The Selfish Gene. The second edition, which came out in 1989, had huge readership so that really helped increase my study’s visibility.
It became controversial later on, I think. At least people began openly doubting the story in various ways. In fact, - the first of several – was, in part, written to try to counter all of the claims that had been made. Back then I did think most of the criticisms were wrong, but I didn’t have a whole lot of data so I couldn’t rebut them. I could only say - I don’t think so. In some ways, Gerald really benefitted from the history of this research project. My story got a lot of attention, then people raised various doubts and over time even openly dismissing my findings. I know people who thought I must have been wrong. But now, given the data Gerald has collected, which is an order of magnitude more than I collected, we can show that the claims people were making were not correct. Explaining it all is still a bit of a challenge but I think the essence of the story in that paper is more or less correct.
HS: Recently, you have gone back to studying vampire bats with your graduate student , but for a long time you had moved away from studying them. Why was that?
GW: Well, when I finished my PhD I had just spent 26 months of a five year period living largely by myself in the tropics, with not a lot of intellectual interaction. At that point, I decided that this was not a good strategy in the long-term. I didn’t want to be like a primatologist spending years on end studying one species. I wanted to do more than that, and I wanted something that could be done in the lab and over shorter time periods – weeks or months, instead of years. So I went and did this postdoc. on Drosophila in the UK, then did a project on mice, before I finally got a job here at Maryland. Then I actually started a new bat project, but instead of vampire bats I started working on some North American bats. At that time, I think I also felt that I had got, kind of, the most interesting story out of the vampire bat, given the technology at that time, and doing anything more would have been very hard. The next step required a captive colony, for which I didn’t have the resources then. So I, instead, worked on various other bat species, on which interesting work could still be done in the field. Then Gerry approached me. He had done some bat work – in fact on vampire bats - as an undergrad at Cornell. Then he did a master’s with working on one of the bird-feeding vampire bats. He contacted me about coming to do essentially a follow up of my PhD work. He really had a very clear idea of what he wanted to do. He had read my papers very thoroughly, knew exactly what I’d found and was aware of the debate over the merits. He loved the story and really wanted to confirm the story. Although, interestingly, when he first attempted to do studies in Trinidad - catch bats, bring them into captivity, and try to get them to feed each other, the bats just wouldn’t feed each other. He actually sent me emails and asked whether I had ever seen them feeding each other! He was seeing them licking at each other’s faces, but they weren’t doing anything he could interpret as feeding. I told him - it is pretty obvious when you see it and maybe your bats are just not used to each other. Sure enough when he saw them in a place where the bats had been together in captivity for a long time he saw them feeding each other a lot - just like what I saw -
But he did see some things that, either I didn’t see, or the bats don’t do in the wild. His bats seemed to actively compete to feed the bat that had been starved, while my starved bats would beg from other bats. My interpretation is that it’s probably because his bats are kept in captivity and they are fed all they need all the time, whereas the bats in the wild, most of the time, don’t have a lot to spare. So it makes sense to me that they would be more picky about who they would share with.
HS: Was handling the bats difficult?
GW: Vampire bats are interesting in that they become accustomed to handling pretty readily. The bats that I kept were very used to being handled and didn’t struggle. In fact, even the bats in the wild, because I caught them repeatedly, often would not struggle. The first time you catch a vampire bat though, it will attempt to bite you and try to get away. And because they can run, they are a little trickier to catch than most other bats. Other bats, you pin their wings back, you have them under control. They can’t do much. But vampire bats, because they can push with their feet, will just squirm out of your hand. But you just get used to it. And I had no difficulty handling and I don’t think Gerald had any difficult either.
HS: How did you catch them? Could you just reach out and catch them when they are roosting, or did you have to use nests?
GW: I learned that if you reach up and catch them while they’re roosting you will cause them to abandon the roost. The only time I would do that is when the adults had left the roost and the young were left behind. During the first several weeks after birth the pups can’t fly, so I would go in the roost, catch the young bats, band them and put them back. That also allowed me to figure out who the moms were because I could see who they would nurse. But adults I would net. Vampire bats are very smart and will learn very quickly where you put the net. So you have to constantly be creative in how you catch them. It was always a struggle, trying to be one step ahead of them.
HS: Where did you do the experiments?
GW: I did this work at a cattle ranch, which had a little sort of tourist area - – and small houses. Some of houses were duplexes and some were standalone. I was in a duplex that had two bedrooms, a bathroom, and a dining room with kitchen. I converted one of the bedrooms into an observation room where I kept the bats in a cage I made with a glass front so I could observe the bats. It had openings so that I could take out individual bats and replace them, and I could weigh them before and after I feed them. It also had feeding tubes which allowed me to feed each bat separately. That allowed me to know how much blood each bat took.
HS: You said that Nature initially rejected and asked you to resubmit. Was the final published version very different from the first draft submitted? Do you remember the big changes that Nature wanted?
GW: That’s a good question. I’d have to go back and check to be sure but I think that the figure with the weight loss - the one with the exponential decay curve - might have been a single panel in the original draft. I think, based on the comments, I broke it out into two - one with the data and one with the expectation. I suspect I still have the original version of this paper in hard copy somewhere.
HS: What kind of an impact do you think this paper has had on your career?
GW: It’s pretty enormous, there is no question. I’m trying to think of the timing – when I applied for the postdoc. in Sussex, either the paper was already published or I had submitted it. I’m not sure. But I knew the results. And once the paper came out, the fact that it was in Nature, and that I was the only author, made a lot of difference.
HS: That is quite rare isn’t it, a single-author paper in Nature?
GW: I guess. I don’t know what the numbers are, but it will be interesting to find out. Maybe it is more uncommon now than it was then. But in terms of citations, even now, this paper gets cited more, every year, than any paper I write. 33 years after I wrote it! I find it incredible that people still refer to it so much. But if you look at the papers that cite it, you realise that a lot of people cite it without having read it, because it’s not directly relevant. For example, they are talking about reciprocity in a different context and they just want to cite an animal example. This paper has kind of become the go-to example for reciprocity in animals.
HS: Is that what it is cited for mostly these days – a non-human example for reciprocity?
GW: That would be my guess. I got to know a lot of people studying reciprocity because of that paper. In 1986, there was a meeting in UCLA, to commemorate 15 years after Trivers’s paper. Trivers was invited, Frans de Waal was invited and a bunch of other people. By that time there were a variety of studies on other taxa where people had claimed reciprocity. We all came and we gave our stories. There was also an. But now, a lot of the examples presented at that meeting have kind of fallen by the wayside, although I think the primate literature has expanded enormously. Today, in the non-primate literature it’s harder to find convincing examples.
HS: Did this paper have a role in you getting the current position you hold?
GW: Oh yes. I am sure it did not go unnoticed when I was hired and then I was promoted with tenure a year earlier than I need to have been. It is all kind of connected is - having done that study and talking about that work at different places - at Oxford, at meetings in Scotland and Australia. People in the bat community got to know of me. Ted Fleming, for example, who I got to know while I was doing this work, was, I think, one of the people who wrote letters for my promotion. Also because it was published in Nature, it just got more attention. I was lucky. The timing was perfect in some ways.
HS: You spoke about the coverage this got in the popular press, including . At the time this was published, was the idea of reciprocal altruism known and being discussed by people outside academia?
GW: Well, the prisoner’s dilemma is, as you know, something that crosses disciplines. People in biology, political science and economics are familiar with it. Whether it went beyond that into the non-academic community, I am not sure. But as an example, I was once asked to respond to questions, on this work, posed by grammar school kids in Japan. A woman I know, who studies bats in japan, translated, I think the Scientific American article, into Japanese. These school kids read that article and the school teacher wrote to me asking if I would answer some questions. It was great to speak to those kids. It is not usual to be able to reach that level of audience. It is partly also the mystique around the vampire bat.
HS: In the 32 years since it was published, have you ever gone back to reading this paper?
GW: You know, if we had done this interview 10 years ago, my answer to this question would have been ‘no’. But recently I have had to read it because of Gerald Carter. Because of his work and the need to put his work in the context of my earlier work, I’ve had to look at this multiple times. In fact I have even pulled out all my old field notebooks with all the data and run network analyses on my associations to see what they look like.
HS: When you read the paper now, what about it strikes you the most? Has your writing style changed substantially since then?
GW: Well I probably never spent as much time writing a paper as I did with this. It took a lot of time and effort and a lot of rewriting. When I read it now it seems pretty dense. Every sentence has information in it. Nowadays, for a paper like this there would be a 100 page supplement that would have all the data and all kinds of other analyses. So that’s one difference. I think the writing style also seems somewhat different although it is hard for me to pinpoint exactly where the difference lies. It’s very natural history-focussed at the beginning, but there are three clear predictions after that, followed by the evidence in support of them. Even if I wrote it now, that would still be the way I would want to do it. You know, come up with such and such predictions and then come up with the evidence to support them. But what strikes me when I read this paper now is just how little data there really is! And how much effort went in to getting those data. It was Herculean.
HS: If you were to repeat the study today, would you do it differently, especially given the availability of new technology and the theoretical advances in your field?
GW: Even then I had many other ideas that didn’t all work. I made radio-transmitters with little waistbands that were designed so that when the waistband stretched the pulse rate of the radio would change. Using this I hoped to get remote data on how much the bat had fed. I made the device, put it on the bat and it worked. But, I also had to make holes in the membranes of the bat’s wing to get the waistband on and it would still slide off the bat’s belly when it fed. I eventually decided it was too intrusive.
The main advantage today would be better video technology, but doing what I did would still be hard – you still need to get into the hollow, and keep the bats in view, which often requires squirming around inside the roost. So, I think what Gerald Carter did – work with a good captive population – is the way forward. He is a postdoc down at Smithsonian in Panama now, where he has created his own captive group. That is definitely the way to do it because you can observe them without any interference and manipulate them in any number of ways. It’s just that once you bring them into captivity you need to keep in mind that feeding them may influence them in more ways than you expect.
HS: And of course, it is also easier to measure relatedness today.
GW: Oh yes. He was able to genotype 30 microsatellite loci while I did only seven allozyme markers, which was actually pretty good. The other thing he was able to do is to record and do playbacks of ultrasonic calls. That technology was not available to me. To record ultrasound you needed a high speed tape recorder, which was a great big bulky thing with big batteries and very expensive. I didn’t have one of those. Today, doing the same is still not cheap, but much cheaper than earlier. And less cumbersome.
HS: You said that you went back to your field site 15 years ago. Were the colonies you studied still there?
GW: Yes, there were still bats in the place I worked at, the last time I was there. In a lot of places in Latin America, vampire bats are actively managed or exterminated because they are carriers of diseases of livestock – Rabies and Equine Encephalitis. But in the nature preserves they don’t harm the bats, so they are still reasonably common. Especially if there are livestock in the landscape, they have abundant food and it’s easy for them to survive and do well.
In the specific site I worked at, the bats were there 15 years ago. I tried but I didn’t find any of my banded bats. Some of the trees where I had watched bats had fallen over and no longer had bats, but there were still a couple that did. Vampire bats can easily switch roosts. They are not tied to a place and move around quite often.
HS: Had your field site changed a lot since when you worked there?
GW: Yes. It was no longer an active cattle ranch, though it still had some livestock. I know that it was turned into an ecotourism kind of place and is still called ‘Hacienda La Pacifica'. It is near the town of Cañas in Guanacaste province. The people who owned the cattle ranch were from Switzerland, originally, and they managed the land better for wildlife than most cattle ranchers. They left large tracts of second growth forest and pretty good strips of riparian vegetation along the rivers that bordered their property. It was in this riparian forest primarily where all the hollow trees were.
HS: What would you tell a student who is about to read this paper today?
GW: Well, I think I would encourage him or her to read the more recent papers of my graduate student, as a way to bridge the gap between the 1984 paper and now. I was pleasantly surprised that what Gerald found, with a lot more data, was not all that different from what I had said in the 1984 paper. In the '84 paper, the point I tried to emphasize, which is often not mentioned, is that kinship and association are equally important. Most people focus on the reciprocation, maybe because that’s the more unusual result. Interestingly, in the recent work kinship seems to be less important. Now that may just reflect the composition of the bats in the group. The bats Gerald used came from more than one place, and so many were not closely related. Whereas in my site, bats were probably a lot more closely related. At this point it is an open question - if one were to go to other places and look at vampire bats and survey relatedness what would one see? It might be that what we found was somewhat unusual or maybe it was typical. But one thing that is true is that they don’t need to be related to feed each other. I think that’s pretty clear. I also think that the bats are highly dependent on food-sharing. If they didn’t share food, they couldn’t live for 18 years as some of them do.
HS: Is this your favourite paper?
GW: That’s an interesting question. I guess it is in many ways. If I went through my papers I could tell you why I liked each one of them, but this one is easy to like from many angles. The funny thing is my dissertation had three main papers in it – this one and two, published back to back in Behavioural Ecology and Sociobiology, on social organisation of the vampire bat – and . The latter two are the papers that I had always imagined I would write. The data that I collected was kind of geared towards producing those papers. This paper was an afterthought. I did the experiments and once I realised how the results had turned out I thought – Oh my goodness, I need to flesh this out quickly. So, the time period in which this paper was conceived and finished was much shorter. And, of course, it had much greater impact too.