by Dave Goulson
Given the huge amount of energy that a bumblebee needs to fly, their ability to navigate accurately and swiftly between flower patches and their nest is the key to their survival. They are phenomenally efficient at finding their way back to rewarding flower patches over and over again. In this way the more robust species such as buff-tails can eke out a living in inhospitable, intensively farmed areas where flowers are few and far between. The key to helping our rarer species to thrive is probably simply to add more flower patches to the landscape, making it a little easier for them to find food and keep their nests well provisioned.
CHAPTER SIX
Comfrey and Smelly Feet
Take time to smell the roses and eventually you’ll inhale a bee.
Anon.
In the summer of 1996 I found myself watching bumblebees, sitting amidst a dense patch of nettles and comfrey in the Itchen Valley Country Park, on the northern edge of Southampton. I had managed to get myself appointed as a lecturer in the biology department at Southampton University, and was enjoying for the first time the luxury of being paid to do research on anything I fancied.
The job of university lecturer is a pretty odd one. For a start, an ability to lecture is fairly low on the list of attributes for which the university appointing panels look. The main criterion that is used is the research record of the candidate. This explains why some university lecturers are entirely lacking in even the most basic communication skills, or indeed social skills of any kind. When I was an undergraduate I had one lecturer who, in a manner slightly reminiscent of but much less engaging than that of my old biology teacher, Mr Newton, would deliver an entire one-hour monologue with a pipe firmly clenched between his teeth and his back to the audience so that almost nothing he said was audible. Another appeared to suffer from narcolepsy and would fall asleep mid-lecture, his eyes closing and his head nodding forwards. In the embarrassingly long silences that ensued we would fidget and eye up the door, considering making a break for it, but usually after a little while he would splutter back into life.
To be fair, I also had some fantastic lecturers, but my point is that teaching skills had clearly not been high up the agenda when my lecturers were first interviewed. The situation is much the same today, and so it was in 1994 when I was interviewed at Southampton. I was a spectacularly shy child who had gone to extraordinary lengths when at school to avoid ever having to stand up in front of the class. As an undergraduate I had faked a range of life-threatening illnesses and thereby managed to avoid ever giving any kind of presentation. I was thus hardly ideal material for a job that routinely involved standing up in front of 200 or more students for an hour at a time, but that did not matter.
My interview took place on a lovely sunny day and Southampton somehow managed to impress me. It is not a beautiful city by any stretch of the imagination; in old photographs the city centre looks splendid, crammed with timber-framed pubs and the offices of wealthy shipping merchants, but some drastic remodelling in the 1940s courtesy of the Luftwaffe removed much of its charm. (Hitler again – who would have thought that the extinction of short-haired bumblebees and the demise of Southampton’s ancient city centre could have a common cause.) Nevertheless, the area around the university is green and pleasant, with a very large expanse of common land covered in ponds, woods and grassy meadows, right in the heart of the city. That day, I gave a nervous and very shaky presentation to the staff of the biology department on the research I had done in an earlier postdoc position at Oxford (on the peculiar mating behaviour of death-watch beetles), and I was then grilled on my research plans. Thankfully I can remember none of the details, but the interview did not go well. I did not get the job – I was ranked fifth of the seven people interviewed. I returned to my postdoc in Oxford where I was working on how to kill caterpillars with viruses, a most depressing project. I felt dejected and rejected.
Then, four months later, out of the blue, I received a telephone call from Southampton offering me the job. The four preferred candidates had all either turned down the job, died in mysterious accidents (I have alibis) or were otherwise no longer available, and suddenly I was in the frame. I was delighted. Naively, I failed to negotiate at all on salary, accepting the job before any figure had even been mentioned. I started as a lecturer in January 1995 and I was to spend the next eleven years working there.
Although Southampton may be few people’s idea of an attractive place, Hampshire is a lovely (if rather crowded) county. Just to the west of Southampton is the New Forest, with its mix of ancient woodlands, windswept heaths and quaint chocolate-box villages; a fantastic place for a lover of wildlife. It is a particularly good hunting ground for bush-crickets, for it harbours a number of rare species. On my weekends I spent many hours trying to track down the Roesel’s bush-cricket, a beautiful emerald-green-and-black beast which inhabits the salt marshes on the southern edge of the forest. The males make a high-pitched hum which is almost impossible to pin down, at least to a human, but is presumably effective at attracting females of the species. I became interested in coneheads, slender bush-crickets with rather angular heads (and a great name). There are two species in the UK, short-winged coneheads and long-winged coneheads, but the short-winged ones can sometimes have long wings, just to confuse matters. For unknown reasons both species seemed to be doing very well in the mid 1990s, using their long wings (when they had them) to spread northwards from the south coast. If I had another lifetime to live I might spend it studying crickets.
Just north from Southampton are the South Downs, the chalky hills that run from Hampshire to Kent. Although much of the land is now intensively farmed, there are still fantastic pockets of flower-rich chalk grassland, relics of a time when the Downs were all one vast flowery meadow. One of my favourites remains Yew Tree Hill, a little nature reserve owned by Butterfly Conservation which is awash with wild flowers through the spring and summer, including huge numbers of the beautiful purple spires of pyramidal orchids. I performed many studies of bumblebees there while I was at Southampton. The Downs must once have been the most splendid place.
Running south from there to the sea are many crystal-clear streams, some of them used heavily for rearing watercress, a local speciality which needs unpolluted waters. Several of these streams converge to form the Itchen, which flows down past affluent Winchester into the Solent near Southampton. It is a lovely, clear river, with long green streamers of water crowfoot which in summer end in delicate white blooms just above the rippling surface. Trout and elegant grayling (my favourite fish) sport among the waterweeds, eyeing up the clouds of bright blue damselflies that skim above the water. Half a mile or so before the clear waters of the Itchen merge into the murky tidal soup that is the Solent, they flow through the Itchen Valley Country Park, where I found myself sitting in the summer of 1996, watching bees.
The Itchen Valley Country Park is an area of lowland water meadows, ditches, streams and woodland, hedged in by roads and housing estates. It was not a peaceful spot then and is less so now; there is a constant drone of traffic from the nearby M27 motorway, and planes roar low overhead every few minutes, taking off from the nearby Southampton airport. Nonetheless it has plentiful wildlife, with otters, water voles, and swarms of the gorgeous demoiselle dragonflies along the riverbank. I had noticed that comfrey flowers, both there and elsewhere, are very variable in colour; some are white, others pale mauve, dark purple and just occasionally bright pink. This is quite unusual for a wild flower – most tend to be uniform in colour, more or less. I was curious to find out whether different insects preferred particular colours, and so I had sat down to watch. The visitors to comfrey are almost all bumblebees. Comfrey flowers are quite deep, so that only long-tongued bumblebees such as the garden bumblebee and the common carder can reach the nectar, but I noticed that white-tailed and buff-tailed bumblebees, which are short-tongued, would readily bite a hole in the side of the flower and steal the nectar (something known as nectar robbing). If I listened closely I could actu
ally hear their jaws chomping through the petal. Early bumblebees would also visit the flowers; they are short-tongued but did not bite holes in the flowers. Instead they rely on those already created by the buff- and white-tailed bees. Almost every comfrey flower had a neat hole in the side. All of this was bad news for the flower, for robbing bees do not come into contact with the reproductive parts of the plant and therefore do not pollinate it. Fascinated, I watched for many hours, recording the sequence of visits by each bee to the pink, purple or white flowers. It turned out that none of the bees seemed to care what colour the flowers were, readily flitting from one to another, showing no preferences; not the most exciting result.
I was about to give up on all this when it dawned on me that I had been seeing something else that was very odd, although it had nothing to do with the colour of the flowers. Dozens of bees were buzzing around in the comfrey patches, flitting swiftly from flower to flower. Very often they would fly up to a flower, hover in front of it, but then fly away without landing. They might do this to three or four flowers before finding one that was apparently to their taste; then they would land and feed. What were they doing? I was puzzled. To try to find out, I measured the nectar in these rejected flowers. This is a fiddly business. The usual approach is to use a very narrow glass tube called a capillary tube, about 1 millimetre in diameter. This has to be carefully pushed into the flower in exactly the way in which a bee inserts its tongue, so that the tip of the tube touches the tiny drop of nectar in the nectary at the bottom of the flower. When this happens the nectar is sucked into the tube all on its own, due to something known as capillary action. By measuring how full the tube is with a ruler, one measures the amount of nectar. So I watched bees, and measured the nectar in hundreds of the flowers that they rejected. I did the same with the flowers that they landed on, quickly shooing the bee away after it landed but before it could drink. It turned out that those flowers that were being rejected had less nectar in them than the flowers the bees landed on. Somehow they could tell which flowers had the most rewards, and were avoiding landing on those that didn’t have much in them. How were they doing it? I was intrigued.
Luckily, at about this time I managed to get funding from the university to support my first PhD student. By coincidence, Jane Stout, a recent graduate with a first-class degree in Environmental Sciences from Southampton, had just returned from an expedition in Tanzania and was looking for a job, so I persuaded her to have a bash at doing a PhD. Jane and I were to spend much of the next four years trying to work out exactly what these bees were doing.
Bees that are collecting pollen can often see which flowers have most reward, because in many flowers the anthers that produce the pollen are visible from a distance. Pollen is usually brightly coloured: yellow, orange, white or purple. Look at a rose, geranium or bramble, and to the human eye it is obvious that some flowers have more pollen than others (the ones with little pollen usually being ones which have recently been stripped by a bee). Bees have pretty sharp eyesight, and we found that pollen-collecting bees do indeed quickly learn to spot the most rewarding flowers. But the nectar in comfrey is hidden in the bottom of the flower (or, rather, in the top of the flower, since the flower hangs down like a bell), and in any case is a colourless liquid. How could bees tell which flowers had most nectar in them? In fact, it turned out that they couldn’t. We took flowers that had just been emptied by a visiting bee, and added more nectar to them (having carefully sucked it out of another flower). These flowers were still rejected by passing bees. We also covered flowers with netting for several hours to keep bees away and allow them to fill with nectar, and then artificially sucked the nectar out with capillary tubes. Bees readily landed on these empty flowers, only to be disappointed on finding nothing inside. So it seemed that the bees weren’t able to tell how much nectar was in flowers, yet in natural situations without scientists messing everything up for them they somehow only landed on the full ones. How on earth were they doing it?
After a lot of pondering, I realised that the bees must somehow be able to tell which flowers had recently been visited by another bee. Regardless of whether it was full or empty, bees would not land upon a flower recently visited by another. Similarly, even if they were empty, bees readily landed on flowers that had been covered in netting to keep them away. So it had to be something to do with the bees visiting the flowers. But what? Bees didn’t leave behind any visible mark when they visited a flower, unless they bit a robbing hole, and almost every flower already had one of those so that couldn’t be the clue. Perhaps they could smell where other bees had previously been? We tried washing the feet of each bee with a tiny drop of solvent, then pipetting the drops on to flowers – and hey presto, those flowers were rejected by other bees. If we pipetted only solvent on to the flowers, there was no effect. So it turns out that bees leave behind a smelly footprint when they land, which any subsequent passers-by can detect, alerting them to the fact that the flower has recently been emptied.
So why do bees have such smelly feet? In fact all insects are covered in an oily liquid that helps to keep them waterproof. It is not just their feet; their entire bodies are bathed in a thin layer of oil. Each insect species has its own particular blend of these oily hydrocarbons, leaving tiny traces on anything they touch. The antennae of insects are finely tuned to detecting these traces, so that they can readily smell just a few molecules in the air around a flower, and this warns them that it is not worth the bother. By saving time that would be wasted climbing inside empty flowers, bumblebees can gather more nectar per hour and their nest can grow more quickly.
Of course flowers refill with nectar and using smelly footprints to detect empty flowers would only work if the footprint wore off. This seemed to be exactly what happens. We recorded how long a comfrey flower remains repellent to passing bees after it has been visited, and found that the effect seems to wear off after about forty minutes. We then carefully measured how quickly the flowers refilled with nectar, and found that they took between forty minutes and an hour to refill. In other words there was a pretty good match between when bees would visit a flower and when it was likely to be full, or nearly so.
With our feeble sense of smell, this all seems terribly impressive. However, there is more. It turns out that different flower species refill with nectar at different rates. For example, borage produces nectar very fast, whereas comfrey is middling, and bird’s-foot trefoil is very slow. When feeding on borage, bees start revisiting a flower just two minutes after the previous visit, and again this roughly corresponds with the time it takes to refill. On comfrey, as I mentioned, bees revisit flowers after about forty minutes, while on bird’s-foot trefoil a flower seems to remain repellent for at least 24 hours.10 Yet the footprints are the same.
How does this work? It seems that bumblebees are able to tell how old a footprint is, perhaps by the strength of the smell, and that they learn an appropriate threshold for any particular flower. This is helped by the fact that individual bees tend to specialise for several days at a time, and sometimes for their entire life, in visiting just one flower species over and over again, gaining a lot of experience. So a bee visiting borage quickly learns to ignore all but the freshest smelly footprint, while a bee visiting bird’s-foot trefoil learns that a flower that smells even faintly of another bee’s feet is best avoided.
We later looked at other insects, and found that this system seems to work across species. Both honeybees and all of the bumblebee species that we studied seem to use footprint smells to judge which flowers to visit, and are able to recognise the footprints left by other species just as well as their own. This of course makes sense, for it does not matter who has emptied a flower – it is still empty. Bees even seem able to tell when a flower has been visited by a hoverfly.
It is probable that without their ability to detect and avoid empty flowers, bumblebees would struggle to survive. In our comfrey patches in the Itchen Valley Country Park there were hundreds of bumblebees fo
raging together, and most of the flowers were empty or nearly so at any one time. Landing, pushing her tongue into the flower and then taking off again all takes time and energy for a hungry bee. Even saving a fraction of a second can, cumulatively, pay huge dividends, for each bee must find tens of thousands of full flowers per day if she is going to fuel all of her flight and bring back a net return of nectar for the nest.
This aspect of bee behaviour was tremendous fun to explore over a number of years, but it turned out that we got one part of it wrong. Jane and I stopped working on bees’ smelly footprints in about 2000, conceitedly thinking that we had pretty much wrapped up most of the interesting angles on this. I started working more on the ecology of rare bees, and Jane got herself a lectureship at Trinity College Dublin, where she has since made a name for herself studying the pollination of invasive weeds such as rhododendron. In 2005 I found myself at a conference in St Petersburg chatting to Thomas Eltz from Dusseldorf University. He had been analysing the speed at which the chemicals in bee footprints evaporated from flowers, and had found that it was far too slow to fit with our explanation. The compounds are very large and not very volatile at all, so that they remain on the flower more or less indefinitely. In fact he found that flowers accumulate chemicals from successive bee visits so that, with careful analysis, the flower can provide a record of all the insects that have visited it during its life.
