A Sting in the Tale

by Dave Goulson

  One odd consequence of males being haploid is that they do not need the process of meiosis to produce haploid gametes; all of their cells are haploid, and hence all of their sperm carry exactly the same set of genes.

  My apologies if you are on the verge of falling asleep. I teach this stuff every year to the third-year students at Stirling, and every year I notice that at least half the class have tuned out within five minutes, no matter how much I jump about and try to make it sound exciting. The importance of all this is that it results in some very odd patterns of relatedness. Daughters get one copy of each chromosome from their mother and one from their father, so they are 50 per cent related to their mother, and she to them. Sons get one copy of each of their mother’s chromosomes, so they carry 50 per cent of her genes. Sisters must share identical genes from their father’s side, and on average share 50 per cent of the genes they get from their mother; hence overall they are 75 per cent related. This is a crucial point – sister bees are more closely related to each other than they are to either their mother or their own offspring. Another strange quirk of this system is that a father is 100 per cent related to his daughters (she has all of his genes), but she is only 50 per cent related to him (half of her genes came from her mother).

  If by now you are utterly confused, as I was when I first tried to get to grips with this, don’t worry. All that you really need to remember is this: a female bumblebee is 50 per cent related to her daughters and sons, but 75 per cent related to her sisters. Now, why does any of this matter?

  I started this slightly tedious discourse on bumblebee genetics by saying that parents look after their offspring because their offspring carry their genes. Now, put yourself in the (six very small) shoes of a worker bee in a bumblebee nest. She could try to lay her own eggs. Because she has not mated, these will be sons, which carry 50 per cent of her genes. Even supposing she had been able to mate and could produce daughters, they will also carry only 50 per cent of her genes. Alternatively, she could help to rear sisters, which carry 75 per cent of her genes. So all else being equal, the best way to increase the number of copies of her genes is to rear sisters rather than her own offspring. This, in essence, is why highly social behaviour has become common in the Hymenoptera, but is rather rare in most other organisms; their odd genetics have predisposed daughters to help their mother rear their sisters rather than trying to reproduce themselves. An ant, wasp or bee nest is a vast, tightly knit group of closely related sisters helping their mother to produce more and more sisters.

  This, hopefully, explains why a bumblebee nest is so harmonious; so long as they are rearing their sisters, the worker bees should be content. They have no incentive to try to have their own offspring. The problem comes when it is time to produce sons. Somebody has to produce them; in bumblebees, the nest will die in the winter, so the only option is to make new queens and for the males to mate with them before summer’s end, so that the mated queens can survive the winter. Hence, in high summer, the queen bee starts laying both fertilised (female) and unfertilised (male) eggs. Up to this point she has been releasing a pheromone signal, instructing her female offspring to develop as workers rather than queens. At about the time that she starts laying male eggs, she switches off this signal. The goal of the queen through the spring and early summer has been to build up a big workforce. Now, her aim is to use this workforce to produce as many new daughter queens and sons as she can, in the hope of ensuring that she leaves descendants in the following year (I should stress that I am not implying that the queen has actually thought this all through).

  This is all very well for the queen: she is equally related to her sons and to her daughters (50 per cent), so she is content to produce both. But rearing brothers is not such an attractive proposition for her daughter workers. Since their brothers have no father, they are equivalent to half-brothers, and so they have only 25 per cent of their sisters’ genes. If the nest must produce males, the workers would rather rear their own sons (to which they are 50 per cent related) than their brothers. Although the worker bees are physically unable to mate, they have perfectly functional ovaries, and can lay unfertilised eggs – which will develop into sons.

  This conflict over who produces the male offspring leads to chaos. It takes a few days for the workers to detect that there are male grubs in the nest, at which point they start trying to lay their own eggs. The queen cannot tolerate this treachery, and so she sets about eating all of the workers’ eggs as quickly as they are laid, thereby consuming her own grandchildren. The queen would much prefer the nest to rear her sons, to whom she is 50 per cent related, rather than her grandsons, to whom she is 25 per cent related. If she catches one of her daughters in the act of laying eggs she will attack and bite her repeatedly; being larger and stronger than her worker daughters she wins the fight easily enough, and then consumes the eggs. However, she is heavily outnumbered. She may win the first fight, and the second and third, but her nest may contain hundreds of daughters, and she cannot bully them all. Now the daughters retaliate, eating the queen’s eggs – their baby brothers – and anarchy ensues.

  Until recently it would have been difficult to work out who won these battles, but these days it is straightforward to use genetic markers to identify the mothers of males in a nest. Steph has done this with some of her buff-tail nests, spending hours in the lab genotyping thousands of male bees, and she has found that the vast majority of the males are sons of the queen. Other researchers have looked at other bumblebee species and generally found that workers manage to produce at most 10 per cent of males. It seems that, despite being heavily outnumbered, the queen is able to hang on to power for long enough to ensure her own reproductive success.17 Nevertheless, the running battles between the queen and her many workers take their toll, and her condition deteriorates – her wings become frayed and her fur thins as she accumulates injuries. From this stage on the nest’s days are numbered, for no new workers are being produced, and with some of the existing workers battling over laying eggs in the nest, the incoming food supply dwindles. Sometimes the queen is even killed by her daughters, whilst at other times the food supply runs out and the remaining workers simply wander off, leaving their listless mother to expire amongst the ruins of her nest.

  This inexorable process is not as sad as it seems. From the point of view of the queen and the genes that constructed her, even if she is finally killed by the daughters that she patiently reared, her life will have been a success if her genes persist in new queens, by now safely hibernating underground, or as sperm stored inside such queens.

  CHAPTER TEN

  Cuckoo Bumblebees

  The Cuckoo comes in April

  She kills a Queen in May

  She enslaves her brood

  To gather up food

  And in July she dies away.

  Anon.

  Most people are familiar with the cuckoo’s nefarious habits. Much as it may seem anthropomorphic to impose human values on a bird, it is hard not to find the cuckoo’s habit of laying eggs in the nests of others rather underhand and distasteful, particularly since the baby cuckoo goes on to slaughter its nest-mates, pushing the helpless chicks out.

  Far less well known, however, is the fact that many other birds, such as moorhens and various ducks, will also routinely sneak their own eggs into the nests of other birds. They behave much like a cuckoo, waiting until the nest owner is absent before swiftly depositing one or two eggs. The unsuspecting owner returns and subsequently spends time and energy looking after extra offspring not its own. This sneaky strategy’s advantage is that it enables the interloper to produce more offspring than it could look after by itself.

  Remarkable recent work by Carlos Lopez-Vaamonde at the Institute of Zoology in Regent’s Park, London, has shown that bumblebee workers do something very similar. Lopez-Vaamonde was using DNA markers to measure how many males within the nests of buff-tailed bumblebees were produced by workers versus their queen. These nests were housed in a lab
oratory high up in a building, but the bees were allowed to forage freely through tubes connecting them to the outside world. He found that workers produced only 2.2 per cent of all males within their nest, with the queen producing 95.7 per cent of males. The really interesting result was that 2.1 per cent of males in these nests were not genetically related to either the queen or her daughters, but were the sons of workers from other nests. As most of the egg-laying workers in the experiment were from other experimental nests, it was easy enough to deduce that bees might have become confused and accidentally flown down the wrong tube and hence ended up in the wrong nest. After all, the nests in the experiment were all very close together, and the entrance tubes probably all looked very similar. So far, so good, but what was really remarkable was that some of the workers laying eggs in these nests had come from wild nests somewhere out in Regent’s Park or in the gardens nearby. That these worker bees had somehow found their way up to the top of a building and into the experimental nests could surely not have happened by chance.

  So what were these bees doing? Remember that, in her own nest, it is not in a worker’s interest to lay (male) eggs so long as her mother is laying female eggs, since she is more closely related to her sisters than to her sons. This argument assumes, reasonably, that the nest has finite resources (i.e. food) and cannot rear unlimited numbers of both. But if a worker can get into a nest of entirely unrelated bees, she should not care that any eggs she lays might be reared in place of the queen’s offspring. Any reproduction she can get away with in this context is a bonus, increasing the genes she passes on to the next generation. We do not yet know how common this is in more natural situations, but a recent study of the Japanese bumblebee, Bombus deuteronymus, found unrelated workers in three of eleven wild nests studied, with these workers producing 19 per cent of the males from these three colonies, so it is clearly not just confined to buff-tailed bumblebees.

  A related tactic is also exhibited by bumblebee queens. Some queens emerge from hibernation later than others, even within the same species. Perhaps they choose a particularly shady spot to hibernate, or manage to burrow deeper into the earth where it is cooler. Or perhaps they are just naturally late risers. Whatever the reason, by the time they emerge from hibernation many of the best nest sites have already been taken. One can imagine one of these late queens, repeatedly exploring promising-looking holes in the ground, and each time finding another queen already in residence. As each day passes without her starting her own nest, the season slips away. It takes time to build up a large nest and produce lots of daughter queens and sons, so if she starts her nest too late it is unlikely to be very successful.

  In these circumstances, it is common for the queen to attack. If she can kill the resident queen, then she can claim control of the nest site and also take over the resident queen’s brood. You may wonder why she should wish to look after another queen’s offspring. In most organisms this would be a very silly strategy, but not so in social insects. The brood are destined to become workers, and they will work just as hard for their adoptive parent as they would for their mother, not being able to tell the difference. A similar strategy is used by Australian choughs (birds of the crow family), which live in family groups with just one breeding pair and lots of younger helpers. These groups will readily kidnap half-grown birds from other groups, incorporating them into their team of helpers and thus improving their chances of rearing more offspring.

  In the case of bumblebees, by killing the resident queen the intruder is saving herself all the hassle of looking after a colony in its very early stages, albeit at the cost of having to engage in a fight to the death against an individual of similar size and strength. We do not know how common such nest usurpation is, but there are accounts of bumblebee nests being excavated to find as many as twenty dead queens inside, and one live one. Whether all of these dead queens were failed usurpers, or whether the nest had been successively taken over by twenty different queens in succession is impossible to say (at least without DNA fingerprinting the queens and workers, which has not yet been done).

  Once a nest has adult workers they should help their queen to repel or kill a usurper, since if their mother dies they are doomed to slavery. For this reason it seems likely that usurping is harder the larger the nest, but on the other hand the prize to be won also becomes greater.

  Such usurpation can take place between different bumblebee species, but generally only between closely related species. Buff-tailed bumblebees, for instance, will often try to invade nests of white-tailed bumblebees, but rarely the other way round since the white-tailed bumblebee queens tend to emerge a little earlier than buff-tails. Similarly, in Arctic North America, Bombus hyperboreus frequently usurps Bombus polaris – and because the Arctic season is so short, the usurping queen does not rear any workers of her own, but instead only new queens and males.

  One group of bumblebees, known as cuckoo bumblebees, have become specialists in this tactic, entirely giving up their social lifestyle in favour of life as specialist assassins. There are six species within the UK, all belonging to the same genus as the ‘true’ bumblebees, Bombus. This means that they all have a common ancestor, and would once have all had a similar life cycle, probably similar to most ‘true’ bumblebees today. But at some point in their evolutionary past, the ancestor of the cuckoo bumblebees evolved down a different route. It is easy to imagine how it happened, and it presumably began as opportunistic usurping, with one late-emerging species often trying to usurp queens of a related, earlier-emerging species. If the likelihood of success in founding a nest by the conventional route was significantly lower than the odds of successfully usurping a queen from an existing nest, then over time the usurping species may have specialised. So doing, they opened up the possibility of evolving physical characteristics to make usurping easier. As a result, these cuckoo bumblebees tend to be larger and have a thicker external skeleton than the ‘true’ bumblebees, which presumably makes it harder for the resident queen or her workers to sting them to death. It is certainly harder to push a pin through the thorax of a cuckoo bumblebee compared to a ‘true’ bumblebee.18 Cuckoos can also be recognised by the lack of pollen baskets in the females; they don’t need them since they don’t do any foraging for the nest.

  We don’t know for sure how cuckoo bees find bumblebee nests to attack, but it must surely be by smell. When keeping artificially reared nests in boxes outdoors it is common to see cuckoo bees flying and walking around searching for an entrance, so they can clearly tell that there is a nest close by. Having found the entrance, the cuckoo bee barges past any workers that get in her way and attacks the resident queen. The queen will usually fight to the death, and with the aid of her workers she may sometimes succeed in killing the intruder. Should she be killed, however, the cuckoo bee will take her place. Occasionally the resident queen will even acquiesce rather than fighting to the death, become subservient to the cuckoo bee and behave like one of the workers. In either case the cuckoo bee will lay eggs, and the bumblebee workers care for them as they would their own. The cuckoo may also eat any eggs or young larvae in the nest, but tends to leave older larvae to develop into workers which will help care for her own offspring. Moreover, cuckoo bees do not produce their own workers, so the female is not, strictly speaking, a queen. All of her eggs develop into fertile offspring, either males or females. Having taken over a nest, the cuckoo bee queen will continue to lay eggs until the workforce she has coerced into her service begins to die off. There is no supply of further workers, so once taken over by a cuckoo bee, the nest will not last for long. But it will usually survive long enough to produce more cuckoo bees, so continuing the cycle.

  The inherited workers continue to work for their new mistress presumably because they have few other options. It is said that they often try to lay eggs, but the cuckoo will chase and bite workers that she finds attempting to do so (just as their mother would have done if she were alive), in a largely successful attempt to keep order
. It would be interesting to see whether these enslaved workers are more prone to drifting off to other nests of their own species to try laying eggs there. Presumably this would be in their interests since otherwise they will spend the rest of their days rearing offspring of an entirely different species.

  Although cuckoo bumblebees all have a common ancestor, there are now thirty or so species in the world, each specialised to some degree on a particular host. The commonest cuckoo bee in England is usually the southern cuckoo, which targets buff-tailed bumblebees (it presumably being no coincidence that buff-tails are the commonest ‘true’ bumblebee species). In most respects the life cycle of cuckoo bumblebees is rather similar to that of their hosts. Mating occurs in mid- to late summer, and only the females will hibernate. Males tend to be much more common than females, and can be very abundant – sometimes the commonest bumblebees to be seen – when feeding sluggishly on flower heads of thistles, knapweeds and bramble. Interestingly, the cuckoo species often have a very similar colouration to their hosts – the hill cuckoo, for example, is black with a red tail, and is superficially very similar to its host species the red-tailed bumblebee. Some years ago, it occurred to me that cuckoo females might also mimic the smell of their host; if they did, it would be less likely that the queen and her workers would sound the alarm and mount their defences. All bees are coated in an oily mix of hydrocarbons, the same compounds which make up the smelly footprints on flowers. The precise mix differs between bee species and also probably differs a little between members of different nests within a species, enabling workers to distinguish nest mates from non-nest mates. I started collecting any female cuckoo bees that I could find and storing them in sealed vials in a deep freeze, with a view to analysing them when I had examples of each species. Unfortunately for me, some cuckoos are rather rare and this progressed slowly. Before I could collect enough to do anything useful, I noticed a new paper by Steve Martin at the University of Sheffield in which he had investigated this in some detail, and shown quite convincingly that cuckoo females do indeed have a smell which closely matches that of their host. My samples are still in the freezer.