by Dave Goulson
Broom has anthers hidden in the bottom of the flower, which swing upwards when a bee of sufficient weight lands, covering her tummy in pollen. Because of this mechanism, broom is largely pollinated by bumblebees, and does not waste its pollen on other less reliable, lightweight pollinators. The trigger plants of Western Australia have spring-loaded anthers, which, when triggered by the arrival of a bee, swing round at high speed like the arm of a mousetrap to smack pollen on to a particular part of the bee. Each species of trigger plants uses a different part of the insect, so their pollens do not get mixed up; some place pollen on the bee’s back, some on her belly, some on her left side, some on her right.
Some bee-pollinated flowers, such as woody nightshade, hide their pollen inside tubular anthers with just a tiny hole at the bottom, from which the pollen can only be extracted by violent shaking, something known as ‘buzz pollination’. Only a few bee species have evolved the ability to do this. Honeybees, for all the sophistication of their colonies and their communication mechanisms, seem not to have mastered the technique. In contrast, bumblebees and some solitary bees are adepts; they grasp the anthers with their jaws and then use their wing muscles to vibrate their body, shaking the whole flower and causing a shower of pollen to fall from the hole in the anthers, which they expertly catch with their hairy legs. Other plants, such as the creeping thistle that spreads in the more fertile parts of my meadow, take the opposite tack; they dispense with subtlety entirely and simply produce heaps of pollen, almost drowning visiting insects in a sea of anthers, so that they become white with sticky pollen. Similarly, large Banksia flowers and many other bird-pollinated flowers in Australia have numerous anthers on long stalks protruding from the flower, so that any bird probing with its bill for nectar has pollen dusted all over its face.
No matter how beautiful the flower, it is not likely to attract insects unless it provides some kind of reward. Many flowers use nectar as their reward; this is simply sugary water, sometimes laced with a few traces of amino acids and other nutrients. Adult insects need fuel to fly, so that they can find mates or places to lay eggs, and nectar is the perfect high-energy drink to keep them in the air – Lucozade for bees. Because plants can create sugars from just water and carbon dioxide by photosynthesis, producing nectar is not enormously costly for them, and some plants produce heaps. Borage and comfrey are both prolifically rich in sweet nectar, and so they are very popular with bees. This presumably gives these plants a distinct advantage in the competition for pollination services.
Other plants produce no nectar, but instead use pollen to entice insects. Pollen is of no interest to some insects, such as most butterflies3 and moths, because they can only cope with liquid diets; they are short-lived creatures, designed to quickly find a mate and reproduce. However, pollen is consumed by some longer-lived insects, including various beetles and hoverflies, and of course by bees, for whom it is a vital protein source for adults and especially for their larvae back in the nest. The protein content of pollens varies greatly, with some plants such as legumes (clovers, vetches, trefoils, peas and beans) producing particularly protein-rich pollen that helps them to attract bees. That is part of the reason why I have been so keen to encourage these plants at Chez Nauche.
It has relatively recently been discovered that the rewards provided by flowers can go beyond the nutritive. Some flowers, including magnolias and some species of lily, generate their own heat. In springtime cuckoo pints, also known as lords and ladies, are common in the hedge bottoms at Chez Nauche. These are a type of arum lily, with peculiar and striking flowers consisting of a purple, poker-shaped vertical rod, called a spadix, partially enclosed in a green hood, the spathe. The male and female parts are clustered at the base of the spadix, beneath a ring of hairs. The spadix produces heat, warming to 15°C above the ambient temperature; they are quite hot to the touch. This helps to evaporate the floral odours, which are distinctly unpleasant and vaguely dungy. Unappealing though they may be to us, they attract pollinators, mainly tiny flies known as owl midges or moth flies. These flies normally breed in decaying organic matter, including sewage; they can even breed in drainpipes and U-bends, and often turn up in the bathroom at Chez Nauche, providing food for the many spiders. These flies are tricked by the plant, for it provides no breeding ground, and they become trapped within the lower part of the spathe, beneath the hairs on the spadix. Once the female parts are pollinated, the plant releases pollen on to the flies, and the hairs wither, allowing the flies to escape. More likely than not they will be tricked again by another cuckoo pint, for they are not the smartest of insects, and so pollen is transferred from flower to flower. In this instance the heat production is purely for the plant’s benefit, and the insects gain no reward whatsoever, but in some plants the heat is used to benefit the insect. For example, flowers of the tropical arum lily Philodendron can be an astonishing 30ºC warmer than the ambient temperature. They attract scarab beetles, which settle, feed and mate upon the lily. The flower folds around the beetles, trapping them in place for a few hours, although in this instance the beetles are presumably happy enough, for they are warm, have plenty of food and spend most of their time copulating and eating at the same time. Once the flower is thoroughly pollinated by all this beetle activity, it releases its own pollen, along with a burst of heat to warm the beetles, before unfolding to allow them to depart, covered in pollen.
Others flowers have evolved structures and mechanisms to help them trap heat from the sun, acting as miniature hothouses. Many bowl-shaped flowers become much warmer than the surrounding air. This effect can be greatly enhanced by heliotropism, whereby the flower slowly rotates through the day so that it is always facing the sun. Probably the best-known example is the sunflower, commonly grown as an arable crop in the Charente. The plants are sown in serried rows across fields of many hectares and, when in bloom in July every single plate-like flower is aligned towards to sun. This makes them particularly photogenic, but it serves a dual purpose for the plant: both warming their pollinators and aiding the germination of pollen grains and fertilisation of the ovules. Heliotropism is particularly common among Arctic and high-alpine plants, where insects may be scarce and temperatures low; for example, flowers of the Arctic poppy keep themselves 6°C warmer that the surrounding air by tracking the sun in this way.
Some fascinating studies carried out by Beverley Glover’s research group at the University of Cambridge have shown that flowers have even evolved special conical cells on the surface of their petals, which focus light like thousands of tiny magnifying glasses on to the cell vacuoles, which are packed with pigment to absorb the heat. These cells are found in Arctic poppies, in spring-flowering crocuses and in snapdragons, which have been a particular focus of their work. The snapdragons have a great advantage as a research tool because a mutant version exists, known as mixta, which lacks the conical cells. This makes it easy to compare the temperature of flowers that are identical in every other way, and to study how bees respond to them. Flowers with conical cells are indeed warmer – up to 8°C warmer than the mixta mutants – and they are preferred by bees. In field experiments with flowers of both types from which the anthers had been removed (to stop self-pollination), the mixta mutants set far fewer seeds because they received fewer visits from bumblebees. However, it has recently become clear that the reason bees prefer flowers with conical cells may have more to do with the extra grip they provide than with their temperature; when landing on the mixta mutants, bees have trouble keeping their footing and slip about, sometimes falling off the flower entirely, whereas the texture provided by the conical cells gives them a firm grip. It is not yet known how common such specialised, grippy petal cells are in wild flowers.
Why might bumblebees prefer warm flowers? Bees need to keep warm to stay in the air. Bumblebees are particularly fat insects with small wings, so they have to beat them 200 times per second to stay aloft. This can only be done if their flight muscles are warm – at least 30°C – which is ha
rd to maintain, particularly in early spring when the ambient temperature may be just above freezing. They have to shiver to get warm enough to take off; once in the air, flying itself creates a lot of heat, and of course their furry coats help to keep the heat in, but nonetheless as soon as they land on a flower to feed they start to cool, and if they sit still for too long they have to start shivering again to warm up, all of which wastes a lot of energy. Hence flowers that are warm are particularly attractive to bees, for this helps to keep them warm while feeding. What is more, a warm flower will have warm nectar: a hot, sweet drink, just the thing for a busy bee on a cold day.
There is a further, alcoholic twist to this story. It has long been known that nectar in plants contains yeasts that are transmitted from flower to flower by pollinators, and which thrive on the sugars, fermenting them to alcohol. It is generally assumed that this does not benefit the plant, and that it may indeed be harmful if too much sugar is used up by the yeast. However, a recent study by Carlos Hererra and Maria Pozo at the Doñana Biological Station in Spain found that the metabolic activity of yeasts in winter-flowering hellebores could raise the flower’s temperature by up to 6°C. Thus the hellebores’ bumblebee pollinators were treated to a warm, sweet, alcoholic drink – glühwein for bees.
Some flowers seem to offer only thermal rewards. Oncocyclus are beautiful, showy purple irises found in the eastern Mediterranean. They seem to have no nectar, and are pollinated entirely by male solitary longhorn bees (which have endearingly enormous antennae) that sleep in the flowers at night. The deep-purple flowers warm up swiftly in the early morning, kick-starting the bees’ day and enabling them, quite literally, to get off to a flying start.
The most perverse strategy adopted by plants in the battle for pollination is by those species that have opted to offer no reward whatsoever. This seems to be particularly common in orchids, a plant family that adopts a range of peculiar and unusual approaches to pollination. I have already mentioned the bee orchids that mimic the shape, texture and smell of the female of the longhorn bees that sleep within Oncocyclus irises. The female bees lack the long antennae, which the males use to locate females. In truth the flowers are a pretty hopeless mimic of the female bee in appearance. Presumably the smell is more convincing, or perhaps male longhorn bees – in common with the males of many animal species – are simply none too discerning when it comes to choosing a partner for a brief liaison. Most flowers produce loose pollen, but orchids package it into balls known as pollinia, each of which has a white stalk with a sticky tip. These are positioned so that the stalk sticks to a visiting insect, which then spends the next few days with the pollinia glued to its face or the top of its head as it travels around. Each flower produces just two pollinia, and they are readily visible, so it is easy to tell whether a flower has had its pollen removed. I have not yet seen longhorn bees at Chez Nauche, and neither have I ever seen a bee orchid flower with the pollinia removed. Through much of northern Europe bee orchids seem to lack their pollinator and rely upon self-pollination to set seed, something that does not bode well for their long-term genetic health and survival.
Mimicking a female bee is a fascinating but understandable approach to achieving pollination. Harder to explain are the many orchid species that offer neither reward nor the false promise of a receptive female. These species rely on naïve pollinators, usually bees, that have not yet learned to avoid them. This may be aided by flowering in early spring, as many nectarless orchids do, for at this time of year there are few other flowers, and queen bumblebees have emerged from a long sleep into a world where all the flowers they encounter are unfamiliar. Some of the later-flowering nectarless orchids appear to mimic rewarding flowers that grow nearby. In alpine meadows in Switzerland I have seen nectarless broad-leaved marsh orchids, Dactylorhiza majalis, growing amongst stands of nectar-rich bistort, which they closely resemble. Male bumblebees seem to be particularly easily duped, perhaps because their mind is on other things, and I have seen some males with dozens of pollinia glued all over their face – so many that they were barely able to see.
In Britain nearly half of the common orchid species and 90 per cent of the rare species appear to offer nothing to pollinators. Darwin speculated that the rarity of many species may be precisely because they offer no reward. This notion is supported by recent studies comparing seed set of orchid species that produce nectar with species that don’t; the flowers of nectar-producing species set seed approximately twice as often as the nectarless species do. This leaves something of a mystery as to why the nectarless species persist with their miserly strategy.
Orchids are not the only flowers to offer no reward. Even in flowers that generally provide nectar, some individual plants produce none. Indeed, this seems to be the norm; some studies suggest that as many as 80 per cent of plant species that are ordinarily regarded as nectar-producing have some nectarless plants. These plants appear to be cheaters, relying on other members of their species to provide the reward. One can measure the amount of nectar in flowers by using a microcapillary tube, a very thin and delicate glass tube that is inserted gently into the nectary of the flower. If nectar is present, it draws it up via capillary action, and the length of the column of nectar visible within the glass tube is a measure of the volume present. Over the years I have spent many hours trying to get nectar out of different flower species, and it is a fiddly and frustrating business; bees are much better at it than humans. At any one time many flowers will be empty because an insect has drained them, but this can be overcome by ‘bagging’ them – enclosing the flower for a few hours in a bag of fine netting to exclude insects, so that the flowers can fill with nectar. I once spent some time studying nectar production in centaury in the meadow at Chez Nauche. Centaury blooms in July, producing small, pink star-shaped flowers. I had become interested in the foraging behaviour of the small, solitary bees that seem to be its main pollinator. No matter how hard I tried, from nearly half the plants I could extract no nectar whatsoever, even when I bagged them overnight.
Producing no nectar is an interesting but risky strategy for an individual plant. It has the clear advantage that the plant saves the energy it would otherwise have put into nectar, but it relies on the presence of enough other plants of the same species that are producing nectar to convince pollinators that the plant species is worth bothering with; or else it requires there to be a plentiful supply of naïve pollinators. The success of the cheating strategy is also very much dependent on the intelligence of the pollinators. Bumblebee workers, for example, are pretty sharp operators and are not fooled easily or for long. They spend their life learning and relearning associations between the colours, shapes and scents of flowers, and rewards. When they first begin foraging they experiment with visiting different flowers, and quickly learn to search out selectively those that are most rewarding. They commonly specialise on a particular flower species, be it red clover, tufted vetch or viper’s bugloss, and become very adept at extracting the rewards swiftly and efficiently – much as we become better at any task with practice. If you follow a bumblebee in a flowery meadow, you will quickly notice that she usually visits the same flower type over and over again, bypassing most other flowers, a behaviour that was familiar to Charles Darwin in 1876 and has become known as ‘floral constancy’:
That insects should visit the flowers of the same species for as long as they can is of great significance to the plant, as it favours cross fertilisation of distinct individuals of the same species; but no one will suppose that insects act in this matter for the good of the plant. The cause probably lies in insects being thus enabled to work quicker; they have just learned how to stand in the best position on the flower, and how far and in what direction to insert their proboscides.
As Darwin observed, this behaviour is exactly what the plants want, for it delivers pollen swiftly and efficiently from one flower to the next, without the pollens from different species getting mixed up and deposited on the wrong stigma. The compl
exity of flowers can be seen as a mechanism to encourage this, a way to manipulate bees into committing themselves to the relationship; if working out how to find and extract the reward from a particular species is difficult, then once the skill is mastered it makes sense to utilise it to the full, not least because competition with other pollinators is likely to be less on such taxing flowers.
All this said, floral constancy only makes sense if the flower that the bee has chosen to specialise on is providing a good reward. Bumblebees constantly reassess their strategy, and occasionally try other flowers that they pass, presumably to check they aren’t missing out on something better. I have my favourite cheeses, but every now and then I try something different that I’ve never tasted before, in case it turns out to be fantastic. Often it is a disappointment – I recently experimented with some Saint Paulin and found it to be tasteless and rubbery, so I’ll not try it again. Similarly, a bumblebee visiting viper’s bugloss might quickly check out a geranium as it passes. If the geranium turns out to have sweeter or more copious nectar than the bugloss, the bee will try a few more; and if the reward is consistent, she will quickly abandon bugloss altogether. If not, she will tend to stick with the bugloss, but will occasionally experiment with new flowers that she happens to encounter. If the bee encounters a run of empty flowers of her favoured plant, she starts spending more time checking out the alternatives. She does the same if a flower becomes hard to find, such as might happen when the favoured flower is beginning to go over.
Bee foraging behaviour can be seen as a trade-off between the benefit of sticking with one flower type that the bee has become skilled in handling and trying to minimise the risk of missing out on something much better. In this way, bees keep tabs on the options available to them, and will swiftly abandon one flower in favour of another if it provides a better reward.