by Colin Tudge
The differences observed in the DNA of existing fig-wasps suggest that they last shared a common ancestor around 90 million years ago. The same is true for existing figs. So that, presumably, is when the ancestor of today’s fig-wasps (who presumably up to then had been making a living in some other way) began to lay its eggs in the developing seeds of the figs’ first ancestor. Ninety million years ago was when the last wave of dinosaurs and the great marine and flying reptiles were still in their prime.
So figs and wasps have depended on each other, and have co-evolved in step – each new kind of fig accompanied by a corresponding, new kind of wasp – for tens of millions of years. Clearly the partnership works very well: both figs and wasps benefit. Clearly (we might suppose) it is in the interests both of fig and wasp to keep the alliance intact. Thus biologists traditionally argued that in such relationships as this, natural selection must favour true and stable mutualism. Peaceable, not to say amiable mutualism has often been portrayed as the natural end-state that would inevitably come about sooner or later. But modern theory suggests that it could be in the short-term interests of either party to cheat; and since natural selection does not look ahead, we might expect that short-term betrayal would indeed take place. Modern studies by Dr Herre’s group and others elsewhere now show that in reality the relationship between figs and wasps has often been reneged upon and otherwise flouted, in various ways and to varying degrees. So biologists must now ask how it is that wasp and fig have served each other so well for so long even though it would pay the wasps to cheat (and the figs too, although probably to a lesser extent).
The relationship between the two can be analysed in the vocabulary of game theory, or indeed of cost–benefit analysis. The bottom line is that the figs and the wasps must each pay a price for the services of the other, but neither can afford to pay too much. The figs sacrifice a lot of their would-be seeds – up to half of the possible seeds in each syconium feed the young wasp larvae. The wasps, on the other hand, seem to exercise restraint – for if they lay their eggs in half the seeds, then why not in all of them? To be sure, this would kill off the figs, the geese that lay the golden eggs. But such things happen in nature – precisely because natural selection does not look ahead, and the long term is sacrificed to the here and now. Alliances of wild creatures are known to break down sometimes, and have led to mutual extinction. In the same way, human societies have often broken down as agreements are reneged upon, to the short-term advantage of some but sometimes to the total destruction of all. Why, then – since it would apparently pay them to cheat – do the wasps exercise restraint?
Just to stir the pot a little more: wasps benefit figs by dispersing their pollen – and as we have seen, the active pollinators are especially adapted to do so. But why? What’s in it for the wasp? In the long term they benefit of course – fertilizing seed to provide a future generation of figs. But since natural selection cannot consider the long term we always have to ask, what is/was the short-term advantage in any particular mode of behaviour? In this case, why does it benefit a pollinating wasp to play an honest game?
In truth, the fig could well have evolved mechanisms to prevent cheating. Perhaps the seeds in any one syconium are not all edible. Plants are remarkable chemists; and virtually all organisms are capable of some degree of polymorphism – producing some offspring of one kind, and some of another. Figs could well produce some would-be seeds in a tasty form, which serve as sacrificial offerings to the essential wasps, and others that the wasps find foul, and leave alone. There is some preliminary evidence to suggest that something like this is happening, but the picture is not yet clear.
Alternatively, the female wasp that enters a new syconium may simply be unable to lay enough eggs to take over all of the flowers. This is eminently possible. As we will see, however, whereas some fruits in some fig species are generally colonized by only a single female wasp, in others – generally the larger fruits – several different wasps invade. Between them, we might suppose, they could colonize all the flowers. So some form of positive repellent does seem likely, to stop them doing so.
But why (in the short term) should the wasps – the active pollinators in particular – go to such lengths to pollinate the fruits they invade? Why not simply pinch the ovules in the syconia they colonize without bothering to fertilize the ones they leave alone? The answer seems to be that any syconia whose ovules remain unfertilized are aborted. No pollination, no nourishment. There is, as the adage has it, no such thing as a free lunch. In short, the fig seems to have rigged the game so that the wasp does gain short-term benefit from playing honestly, and would be punished even in the short term if it cheated.
Such mechanisms seem to have kept the whole system on course. Yet (so the Smithsonian studies have shown) it clearly is not quite so stable as has been supposed. Thus it looked for a long time as if the simplest rule applies: each species of fig has its own particular species of wasp; each wasp is adapted to only one kind of fig. But modern techniques enable biologists to explore the relationships between different creatures by examining their DNA. DNA really is a most obliging molecule. Some bits of it – notably some apparently nonfunctional bits known as ‘micro-satellites’ – change so rapidly that differences between them reveal relationships even within families: who is whose sibling, or parent, or offspring. (These are the kinds of studies used in legal paternity cases.) Other bits of DNA change more slowly, and significant differences between these bits indicate that different individuals belong to different species. This of course is especially useful when two or more different species look very similar. Thus DNA studies in recent years have revealed that various populations of owls, mice, monkeys and bats that were each thought to represent just one species, sometimes should be divided into two or more. When species can’t be told apart except by their DNA they are called ‘cryptic’ species.
By DNA studies, the Smithsonian biologists have shown that fig-wasps include many cryptic species. They even found cryptic species within the ancient wasp genus Tetrapus – suggesting that rule which says ‘one fig: one wasp’ has long been ‘routinely violated’, as Allen Herre puts the matter. We might reasonably guess that when two species of wasp pollinate the same kind of fig, this is simply because some ancestral species of wasp, which served that particular fig, had split to form two species. But DNA studies show that sometimes two different kinds of wasp sharing one fig are not close relatives. This means that one of the two must have come from some other kind of fig. As seems to follow, it also turns out that some kinds of wasp colonize more than one kind of fig. So fig-wasps can be more like honey-bees than had been supposed. When two kinds of wasp meet in one kind of fig, they might in theory hybridize — and DNA studies show that this sometimes happens; although the hybrids do not themselves seem to spread. When one kind of wasp pollinates more than one kind of fig, the figs could be hybridized – and as we have already seen in willows, hawthorns, poplars and many others, including the London plane, hybrid plants of all kinds can and do evolve into new species (and perhaps this is one reason why there are now so many different species of fig).
In the Smithsonian’s Panama studies, whenever more than one kind of wasp colonized the same syconia, both wasps may serve as perfectly good pollinators. But this is not always so. A study in Africa has shown that at least in one case, one of the cryptic wasps that colonizes one particular species of fig tree behaves simply as a parasite. It lays its eggs in the flowers and so feeds its young, but it does no pollinating. It is a cheat: an archetypal freeloader. Game theory predicts that freeloader fig-wasps might exist – it’s a possible niche – and so they do. Again we see that the simple one-to-one relationship between figs and wasps, evolved over millions of years into perfect mutualism, is not quite so cosy as it has seemed. The relationship, like all mutualisms, is dynamic. It is always prone to decay.
Often, too, any given fruit may be colonized by more than one foundress from the same species. This raises another se
t of complications — complications which again have been predicted by modern evolutionary theory, and which again (very satisfyingly) have now turned out to be what actually happens.
Let me refer you back (as Perry Mason would say) to an earlier comment: that the proportion of young male wasps born within a given syconium varies from around 5 per cent (one in 20) to around 50 per cent. The preliminary point is that many creatures can adjust the sex ratio of their offspring. Humans cannot do this, but we do not shine at everything. Such adjustment is especially easy for wasps (and bees and ants) because in these insects, the females all develop from fertilized eggs, while the males develop parthenogenetically, from unfertilized eggs. The mother wasp (or bee or ant) keeps the sperm separate from her eggs until the time comes for laying, so she can decide in the light of circumstance whether or not to fertilize them before laying. Again the mechanism seems so subtle that it beggars belief, and yet it is the case.
Theory predicts that when only one foundress colonizes a particular fruit, the ratio of males should be low: one in twenty rather than one in two. But the more foundresses there are, the more we would expect the ratio of males to increase. The point is that each female strives to pass on as high a proportion of her own genes as possible. If all the larvae within any particular fruit are her own offspring, then all her daughters are bound to be mated by her own sons (if they are mated at all). The shortcomings of incest and inbreeding are apparently outweighed by the advantage that the mother thereby passes on her genes both via her sons and via her daughters. So she needs only enough sons to ensure that her daughters are all fertilized – and one son per twenty daughters seems enough. It is good, of course, to focus on daughters because they are the ones that lay the eggs that supply the grandchildren generation.
But if there is more than one foundress per fruit, then the young males find themselves with rivals who are not simply their brothers (who would be genetically very similar) but come from a different lineage (albeit of the same species). In such circumstances, we might suppose that it could pay a female to produce sons exclusively – provided those sons are big and tough enough to mate all the daughters of all the other foundresses. The theory shows, however, that it never pays to produce more sons than daughters. A 50:50 ratio of sons and daughters is the maximum. Again, natural history supports the theory. Dr Herre and his colleagues have shown that as the number of foundresses per fruit rises to about six, so the proportion of males rises to around 50 per cent.
But here comes another twist. The males have one function only: to mate the females. Apart from that, they are a dead loss – both from the wasp’s point of view, and from the fig’s. After all, the fig has to sacrifice a seed for every young wasp that is born. The youngsters that matter to the fig are the females, which fly off to pollinate other figs. As far as the fig is concerned, the fewer males, the better. This, in turn, implies that figs should encourage wasps to enter their syconia one at a time: that they should evolve some limitation on access (and there are many comparable examples in nature). As things are, small syconia are much less likely to attract multiple foundresses than are large syconia — so we would expect natural selection to favour small syconia. In reality, while some fig species do have small syconia, others have larger ones. So why does natural selection ever favour big syconia? This question will be raised twice more as this narrative unfolds, in two quite different contexts. Whichever way you look at it, big syconia seem like bad news. Yet there is an answer, which will be provided later in this chapter. Patience, gentle reader.
As if the game between figs and wasps were not convoluted enough, there enters now a third set of players: parasitic nematode worms.
ENTER THE NEMATODES
It has been suggested that every species of creature on earth above a certain size has its own specialized nematode parasite; and if this were so, it would mean that the total number of species on earth is equal to the number of non-nematodes, times two. Whether this is so or not, it does seem that every species of pollinator wasp does have its very own species of nematode parasite. All the nematode parasites of Panamanian fig-wasps belong to the same genus: Parasitodiplogaster. Since STRI where Dr Herre works is based in Panama, this is the genus they have studied most.
The life-cycle of Parasitodiplogaster nematodes is superimposed on that of the wasps they attack. Not all figs are infested with nematodes, but in those that are, the nematodes will have reached the immature, dispersal stage by the time the young female wasps are emerging from their flowers. The worms then enter the wasp’s body cavity, and begin to consume it from within. Their efforts are not immediately fatal, however, and so their host carries them on to another syconium. When the infested wasp finally dies, generally after laying her eggs in the next syconium, up to twenty or even more adult nematodes crawl from her body, then mate, then lay their eggs. The young nematodes hatch before the young wasps emerge – and so they are ready to invade the young wasps and begin the cycle afresh.
In general, the relationship between parasite and host is as delicate as that between partners in a mutualistic relationship. The aim of the parasite is to grow and reproduce, and for this it must feed upon its host. If it feeds too vigorously, it is liable to kill the host. If it is too decorous in its approach, it loses out to rival parasites who are more vigorous and so breed more quickly. In general, then, it pays parasites to be as vigorous – ‘virulent’ – as possible, but without overdoing it.
Now a further twist. Theory predicts that if nematodes infest wasps that occupy fruits on their own – one wasp per syconium – then they should be less virulent. After all, if they are too virulent, and kill their host wasps, they have no chance at all of being transmitted to a new fruit to lay eggs of their own. But if the nematodes attack wasps that invade fruits more than one at a time, they can afford to be more virulent. It doesn’t matter too much if some of the young host wasps are killed off, since there are liable to be others which are not killed, and will carry the nematodes to pastures new. The Smithsonian scientists found that this prediction stands up. Wasps that invade fruits singly generally manage to fly off to new fruits even when they are infested with nematodes. But in fruits that entertained more than one foundress, a proportion of infested foundresses perish before they leave the syconium of their birth.
Nematodes are clearly bad news for the wasps; and particularly virulent nematodes are bad for the figs, too. After all, the fig has to sacrifice one of its would-be seeds for every wasp that is produced, and the sacrifice is wasted if the wasp then dies from nematode attack. Again, then, it seems that figs would be better off producing syconia that attract only one foundress. Again, small-sized syconia seem advantageous – because, in general, the bigger the syconium, the more foundresses it is liable to attract. So the question is prompted again: why do some figs continue to produce large syconia?
Finally, neither the fig-wasps nor the nematodes have the field to themselves. A whole number of other creatures – including other kinds of wasp – also feed on the syconia, though purely as parasites; offering nothing in return by way of pollination. Particularly intriguing is a group of wasps in the New World that thrust their ovipositors into the syconium from the outside, and lay their eggs in the outer flesh or in the flowers within. The syconium responds by producing a gall – a mass of tissue that may be seen as a defensive response but also provides a very convenient home for the developing wasps. Galls are common in plants, and particularly in trees which live a long time and are attacked many times. A syconium attacked by a gall wasp fails. Its flowers do not get pollinated.
Yet as we have seen, figs normally abort syconia that fail to become fertilized, so the syconia with galls should simply be shed. But they are not. Evidently the gall wasp stops this happening. The fig mobilizes hormones that cause changes in .the stalks of failed syconia, which cause them to fall off – but the wasp apparently produces chemical agents of its own to subvert or otherwise block the fig’s hormonal signal. The wasp, like
a hacker, has cracked the fig’s protective code. Here is a prime example of the kind of ‘arms race’ (another technical term in biology) that we commonly see in nature as predators and prey, hosts and parasites, evolve in each other’s presence. But – as with the figs and the pollinator wasps – we can be sure that the race is not over yet, and indeed will never be over. We should come back in a million years or so and see if the figs have thought of a new way to overcome the gall wasps’ knavishness. Probably not, for various reasons. But it would not be surprising if, at some time in the future, gall wasps had to up the ante once more. Incidentally, the gall wasps are in turn parasitized by smaller wasps. Jonathan Swift’s observation that fleas ‘have smaller fleas to bite ‘em/And so proceed ad infinitum’ seems to apply just as cogently to wasps. There is more.
COOL FIGS AND HOT FIGS
Although figs surely have no love for gall wasps, they do go to great lengths to protect the vital pollinator wasps. In particular – as again revealed by the Smithsonian studies – they maintain a temperature within the syconia that allows the young wasps within to develop.
