Revisiting Kempenaers et al. 1992

In a study published in Nature in 1992, Bart Kempenaers, Geert R. Verheyen, Marleen Van den Broeck, Terry Burke, Christine Van Broeckhoven and André Dhondt showed that female blue tits choose higher-quality males for extra-pair copulations (EPC), providing evidence in support of the genetic quality hypothesis for extra-pair matings. Twenty-five years after the paper was published, I asked Bart Kempenaers about his motivation for doing this work, how this collaboration came about and what we have learnt since about extra-pair paternity.

Citation: Kempenaers, B., Verheyen, G. R., Van den Broeck, M., Burke, T., Van Broeckhoven, C., & Dhondt, A. (1992). Extra-pair paternity results from female preference for high-quality males in the blue tit. Nature, 357(6378), 494.

Date of interview: Questions sent by email on 29th November 2016; responses received by email on 2nd January 2017.

Bart Kempenaers


Hari Sridhar: This paper seems to have formed part of your PhD thesis. Could you tell us what was the motivation behind this specific piece of work, in relation to the rest of your PhD?

Bart Kempenaers: My PhD was about the mating behaviour of blue tits. I was particularly interested in studying social polygyny and behaviour related to extra-pair paternity, including extra-territorial visits, extra-pair courtship and mate guarding.

Illustration of blue tits © Gerald Driessens


HS:  Stepping back a bit, how did you get interested in the topic of mating behaviour?

BK: My advisor, André Dhondt, taught an animal ecology class, and after hearing about and then reading the work of Nick Davies on dunnocks, I knew I wanted to study mating systems. I was fascinated by how much variation there was in mating behaviour between individuals within a population and by the idea that the mating system is an outcome of sexual conflict, that is, the fact that males and females usually have different reproductive interests. And I was in awe of Nick Davies’ clear writing. My advisor’s main interest at the time was intra- and inter-specific competition, and I did my final thesis (sort of a Master’s thesis) on that topic. He wanted me to continue work on this for my PhD, but it did not interest me enough, so I wrote an application for a PhD position on mating behaviour. Although he was sceptical that I would get the money, he was supportive, and I am still grateful to him for that.


HS: This paper has six authors. Could you tell us how this group came together and what each person brought to the study?

BK: Oh yes, it was a great team. André Dhondt I already introduced. He knew Christine Van Broeckhoven, who by then was already quite a famous human geneticist working on the genetic basis of Alzheimer disease. She was well-known at the University of Antwerp because she was a top scientist, but also because she was considered aggressive and short-tempered. In fact, I was scared of her. But our first meeting went well, and she put a wonderful young woman on the project as technical assistant, Marleen Van den Broeck. So, Marleen did all the lab work and I went to visit her often to bring the blood samples and tell her stories from the field. It was one of those lucky collaborations, where we both inspired each other and became good friends. Geert Verheyen was another PhD student in André’s group who went to the lab of Terry Burke, then at Leicester University, where he developed single-locus minisatellite probes for blue tits, a brand new technique at the time, which turned out to be short-lived, because soon microsatellite markers would be discovered. Looking back, I was incredibly lucky that I could work with Christine Van Broeckhoven. She brought DNA fingerprinting to Antwerp, not that long after it had been discovered in 1985 by Alec Jeffreys. I mean, the first papers showing that it could be used in birds were published in Nature in 1987, and the first study on patterns of extra-pair paternity was published in 1990 in Science. It was on red-winged blackbirds and showed that extra-pair paternity could increase the variance in reproductive success among males, but it did not include data on the birds’ behaviour.


HS:  Could you give us a sense of what your daily routine was like when you were doing the field and lab work for this study – did you visit the site daily, did you have help, were you doing lab and fieldwork simultaneously etc.?

BK: Because Marleen did all the lab work, I could focus completely on the fieldwork. I literally spent more than six months a year in the field, mostly on my own. During the breeding season, I was there every day from before dawn at least until the early afternoon doing behavioural observations. I also caught adults with mist nets or in the nest box when they were feeding their offspring and took a blood sample from all of them, and from all offspring. And when I was done in the field, I drove to the University and first thing I dropped off the samples in the genetics lab and chatted with Marleen when she was in her office. Later in the season, during the nestling period, when the offspring needed to be measured and banded and bled, I got some help from André or from his technical assistant Frans Fierens. They usually took over the last nests of the season in late May, because then I left to Alaska for a month to work as a field assistant on mating behaviour of lekking buff-breasted sandpipers with Richard Lanctot, who became a wonderful friend and supportive colleague.


HS:  Do you continue to work in “plot C” till today? When was the last time you visited it? In what ways has it changed since when you worked there for this study? Is the colour-ringed blue tit population still being studied there?

BK: After my PhD, I left for a post-doc in the Netherlands and then to Canada for another post-doc, and I never returned to Antwerp or to plot C. C stands for Calixbergen. It was a private estate near Antwerp owned by the Bracht family. I have not gone back there and I think the population study also stopped soon after I finished, because my advisor also left Antwerp to take up a position at the Lab of Ornithology in Cornell.

Illustration of blue tits © Gerald Driessens


HS: How were the figures for this paper drawn?

BK: I think I made them with an early version of Sigmaplot.
HS: If you don’t mind, I would like to go over the names of people you acknowledge, to find out a little more about who they were, how you knew them and how they helped:

BK: Luc Bijnens had been a PhD student with André working on blue tit song. He followed a year-long special course in biostatistics at the University of Limburg, but was visiting the department often. He always seemed happy to see me, was interested in what I did and generous with his time. I asked him for advice on the statistical analyses.

Alec Jeffreys had discovered DNA fingerprinting at the University of Leicester. I never met him, but Terry Burke was his colleague and Alec had provided minisatellite probes that were then used by Geert Verheyen in Terry’s lab.

Iris Van Pylen helped Marleen in the genetics lab in Antwerp.

Tim Birkhead is a behavioural ecologist who I had met at a student’s conference in Oxford. I visited him at the University of Sheffield and when I gave a talk he was wildly enthusiastic about my work. So, I sent him the manuscript and he commented on it.

Erik Matthysen was a post-doc in the group of André. With his sharp intelligence he was always happy to critically discuss ideas or results and he gave me some feedback on how to present the results.

The Bracht family owned the estate “Calixbergen” where I did all my work. I am extremely grateful to the young baron Bracht, because he made me feel welcome, gave me a key of the gate and let me work in his “backyard” day and night. That was not obvious, because he lived there with his family and in 1978 his father (Baron Charles Bracht) had been kidnapped and murdered.


HS: Do you remember how long it took you to write the paper, and when and where you did most of the writing?

BK: I wrote most of the paper in my office at the University of Antwerp, and some of it at home. Once I had put the data together and analyzed them, and after presenting the story at meetings, the writing didn’t take that long, all in all perhaps a week or two.


HS: Did this paper have a relatively smooth ride through peer-review? Was Nature the first place you submitted this to?

BK: I remember we talked about sending it to Science, but I cannot remember if in the end we did. I do know that the paper initially had only four authors and was first rejected after peer review by Nature. The reviews were critical but also constructive, and we felt we could easily respond to all the comments. One of the reviewers said that we needed to include data on paternity assignment using single-locus probes. This was an interesting comment, because the reviewer indeed knew we had these data. So, we added Figure 3 and two co-authors and then Nature accepted the revised manuscript for publication.


HS: Do you remember how this paper was received when it was published? Did it attract a lot of attention and discussion?

BK: Yes, it certainly attracted a lot of attention, and mostly very positive. It was a bit overwhelming, really. I was completely unprepared for the effects it had on how people approached me and on my career. It was an interesting experience for a young and naïve PhD student.


HS:  This paper has been cited over 600 times. At the time you did the study did you have any inkling how important it would turn out to be? Do you have a sense of what the paper mostly gets cited for?

BK: No, I didn’t realize that at the time. I knew that the study was the first to bring together behavioural observations of individually marked males and females with parentage analysis, and I was excited by the results, because the picture that emerged was that females were in charge. This was interesting, because extra-pair behaviour strongly benefits males, while for females there are some obvious costs and the benefits are much less clear. I had observed how females left their territory to visit neighbouring males apparently in search of extra-pair copulations. There had been a few earlier studies suggesting that females actively seek extra-pair copulations, but these studies did not have parentage analysis to link the behaviour to the outcome in terms of paternity. I think the paper mostly gets cited for the female role in extra-pair mating and as evidence for genetic (good genes) benefits of extra-pair mating.


HS: Did this paper have any kind of direct impact on your career? In what way did this paper influence the future course of your research?

BK: I think it had a strong impact, because all of a sudden I became something of a young star in the field. It clearly helped my career because without it I certainly wouldn’t have had such an easy time obtaining a Canadian post-doctoral fellowship, let alone the positions that followed. It also influenced the future course of my research in the sense that it was the start of an ongoing interest in understanding the evolution of promiscuity in birds and in other animals. I started two more studies on extra-pair paternity in blue tits, one still ongoing. With my first PhD student Kathi Foerster, who is now a professor at the University in Tübingen, we added an interesting twist to the story, suggesting that females that engage in extra-pair copulations might benefit through obtaining good genes as well as increased genetic diversity for their offspring.


HS: Today, 24 years after it was published, would you say the main conclusions, including support for the genetic hypothesis, still hold true, more-or-less?

BK: Yes and no. It has actually been a fascinating example of how science works. Yes, because many of the observations have been confirmed in later studies on different populations of the same (and other) species. For example, there is little doubt now that females play an active role in seeking extra-pair copulations, and it has also been confirmed in blue tits that extra-pair males are larger. However, the conclusion that females seek extra-pair copulations to obtain good genes benefits was probably premature. Initially, we found additional support for this hypothesis. We could show that extra-pair offspring were in better condition and were more likely to survive than their within-pair half-sibs. Most of my colleagues and I were convinced that this was good evidence for genetic effects, because we thought that within- and extra-pair offspring only differed in paternal genes. These offspring indeed share the same environment because they grow up in the same nest and they have the same maternal genes. At the time, there was some discussion about whether males would treat within- and extra-pair offspring differently, but we thought that this was highly unlikely. Another possibility was that extra-pair offspring would do better because the eggs from which they came hatched a bit earlier, and thus the extra-pair young would have a head start in life. I had actually rejected this idea after looking into it during my post-doc at Queen’s University where I worked on extra-pair paternity in tree swallows and eastern bluebirds. However, when it comes to blue tits, who lay large clutches of up to 15 eggs, it turned out I was wrong. Colleagues from the University in Groningen showed in 2009 in a paper published in Current Biology that extra-pair young are much more likely to be among the first laid eggs and because first laid eggs hatch first and generally do better, this alone might explain the extra-pair young’s superior performance. It’s really great work and I am still shaking my head that we did not do it ourselves.


HS: If you were to redo this study today, would you change anything about them?

BK: Well, this was a study that was based on careful observations without having a specific hypothesis in mind. The fact that females seek extra-pair copulations and have some of their young sired by particular males was like a discovery. In general, this reflects the way I do science. Choose an interesting species, mark all individuals and try to learn as much as you can from observing their behaviour. Nowadays, there are more tools available to study behaviour, such as fancy telemetry systems and RFID transponder systems that allow you to “see” behaviour that previously remained hidden or was not tractable on such scale, but other than that, little has changed. And of course from the molecular side, the tools also improved dramatically with the use of microsatellite markers or single-nucleotide polymorphisms, which allow you to assign paternity and measure individual genetic diversity (heterozygosity) with much higher precision.


HS: You say “A question that is left unanswered is how females are able to judge the quality of their mate and that of the neighbours” and “Perhaps females us the condition of a male during the breeding season as a clue to indicate his quality.” Today, do we know more about how females judge male quality?

BK: Well, I don’t know whether they judge “quality”, but what we did learn is that females are more likely to have extra-pair paternity with those males that start to sing earliest in the morning, during the so-called dawn chorus. Usually, these are the older males. Interestingly, we discovered that artificial night lighting can advance the onset of dawn singing and the males in these “light-polluted” territories are much more successful in siring extra-pair offspring, even if they are young. What is still not entirely understood is why some males naturally sing earlier than others, and how that relates to individual quality.


HS: In the 24 years since this paper was published, have you ever read it again? If yes, in what context? What strikes you most about it when you read it now?

BK: Well, I rarely read my papers again unless I want to cite something, and I hadn’t looked at this paper for a long time. But then last year, I gave a talk in my group in which I presented an overview of my work on blue tit mating behaviour. For that talk, I went through the paper and looked at all the results and evaluated them in the light of what came afterwards. What strikes me most now is how small the sample sizes were. Some of the results are not much more than anecdotes, but I still find them interesting, and I also published follow-up work with larger sample sizes.


HS: Would you count this as one of your favourites, among all the papers you have published?

BK: Yes, but at the time, I was more proud of the work I had done on social polygyny, because I had written the papers (1. 2) without the help of my advisor and both were immediately accepted in Animal Behaviour with minor revisions. One of the reviewers of the first paper that came out in 1994 commented that the manuscript was well-written and reminded him or her of the work of Nick Davies. That put a big smile on my face, because Nick was – and still is – my scientific hero. When I look back today, my favourite paper is Lesku et al. 2012, published in Science. In this paper, we report on the discovery that male pectoral sandpipers hardly sleep during the short arctic summer when they compete for access to fertile females, and that their sleep is highly fragmented with average sleep bouts of 15-35 seconds. That study is the outcome of an amazing journey, with fieldwork with a fantastic team, and with other exciting discoveries. It’s simply my perfect mix of great fun and hard work.


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

BK: I would suggest to read it together with some of the key papers that followed in this field, because by putting it into a historical perspective you get a better idea how science really works. There is a tendency nowadays to lament about the fact that science is not reproducible. There are certainly some important issues that are worth discussing in this context, but part of the discussion bores me because it seems to be based either on false ideas about how science actually works or on unrealistic ideas about how it should work. There is no such thing as a perfect study. Studies are always done in a particular context and not always will the results be generalizable across contexts. That is especially true in our field, where we are working within a complex and variable ecological context. I think it is important that students learn, at the earliest possible stage, that science is not about writing the definitive paper on a particular issue, and that research is not about going in a straight line from point A to B. I like the way Stuart Firestein puts it, in his wonderful little book “Ignorance: How it Drives Science”: science is like “looking for a black cat in a dark room without knowing that there is a cat in the room”. So, it’s about being curious about a phenomenon and studying it from this or that angle, through careful observations or experiments. It’s about drawing conclusions that seem most reasonable given your data as well as your knowledge and thoughts and those of others at that time. And it’s about admitting limitations or weaknesses and openly discussing things we still do not know. Things are rarely black-and-white. Coming back to the evolution of extra-pair paternity, there is still evidence in some systems pointing to genetic benefits, but there are also studies that suggest that female extra-pair behaviour might even have evolved in the absence of benefits, simply because of strong selection on male extra-pair behaviour. There is no simple answer, and it remains interesting, so if a student is interested, he or she can still go out there and discover a black cat, or perhaps it will turn out to be a blue bird.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s