Sometimes a picture can tell you a lot about evolution. This particular picture has a story to tell about how two species--in this case a fly and an orchid--can influence each other's evolution. But the story it tells may not be the one you think.

Coevolution, as this process is now called, was one of Darwin's most important insights. Today scientists document coevolution in all sorts of species, from mushroom-farming ants to the microbes in our own gut. But Darwin found inspiration from the insects and flowers he could observe around his own farm in England.

Darwin's thoughts about coevolution began with a simple question: how do flowers have sex? A typical flower grows both male and female sexual organs, but Darwin doubted that a single flower would fertilize itself very often. Flowers, like other organisms, display a lot of variation, and Darwin thought that the only way flowers could vary was if individuals mates, mixing their characters. (Sex turns out not essential for creating variation, but it does do a good job of creating it.) But in order to have sex, plants can't walk around to find a mate. Somehow the pollen of one flower has to get to another. Not just to any flower, moreover, but to a member of its own species.

The random wind might suffice for some plants. But Darwin also knew that bees visited many flowers to gather their nectar. He began to study what happened on those visits. He would watch bees land on scarlet kidney bean plants, for example, and climb up a petal to get to its nectar. The flower's pollen-bearing organs, Darwin found, were located in precisely the right spot to brush pollen onto the back of feeding bees. When the bees traveled to another scarlet kindey bean plant, they unloaded the pollen. The bees depended on the flowers for food, and the flowers depended on the bees for sex. Without each other, they could not survive.

In the Origin of Species, Darwin offered some thoughts on how this sort of partnership between bees and clover could have evolved. Imagine that the flowers are pollinated by other insects, but the insects go extinct in some region. Now all their nectar goes uneaten. Honeybees might visit the flowers sometimes, and variations that allowed them to reach the nectar--a longer tongue-like proboscis, for example, more easily might be favored by natural selection.

Meanwhile, the flowers would be experiencing intense natural seleciton of their own. Without their old pollinators, their chances of producing offspring plummeted. Any variation that would make it easier for honeybees to pollinate them would bring a huge increase in reproductive success. Gradually, the flowers anatomy would come to match that of the honeybees, just as the honeybees were adapting to the flower.

 "Thus I can understand," Darwin wrote, "how a flower and a bee might slowly become, either simultaneously or one after the other, modified and adapted in the most perfect manner to each other, by continued preservation of individuals presenting mutual and slightly favoruable deviations of structure." 

Around the time that Darwin published the Origin of Species, he developed a fondness for orchids. He was not alone; at the time a rising orchid fever was seizing England's upper class. Aristocrats would dispatch explorers to the Amazon or to Madagascar, where they would strip entire hillsides of the rare plants. Some prized specimens sold for hundreds of pounds at auctions in London and Liverpool. If, as many people then believed, the only meaning of natural beauty was as a gift from God, orchids were the most exquisite gifts of all. They could have only one purpose: to please the eye of man.

Darwin had other ideas.

In orchids, he discovered the same evolutionary pressures at work as in other flowers, but the results were supremely baroque and bizarre. Despite the prices orchids might fetch at auction, their beauty did not exist for beauty's sake. It was, Darwin showed, an elaborate means for luring insects into their sex lives. He documented case after case of these adaptations. One species, for example, had its pollen loaded in a crossbow-like structure that bees triggered by walking across a petal.

Darwin described this and many other adaptations in The Various Contrivances by Which British and Foreign Orchids are Fertilized by Insects, and on the Good Effects of Intercrossing. Darwin guided the reader from orchid to orchid, showing how each flower's design was not simply beauty for beauty's sake, but some of nature's most elaborate forms of sex. He showed how orchids were simply highly evolved flowers. All the various parts of ordinary flowers had simply been stretched and twisted and otherwise transmogrified into new structures such as crossbows.

Darwin was so confident that orchids were adapted to their pollinators that he made a bold prediction in his book. He pointed out how many orchids produce their nectar at the bottom of long tubes called nectaries. The insects that feed on them are equipped with tongues that are almost the same length. Short-tongued insects visit flowers with shallow nectaries, and long-tongued insects visit deep nectaries. In every case, the insect has to press its head against the flower to reach the bottom of the tube. The orchid's pollen is invariably positioned in a place where it can stick to the insect's head while it drinks.

Darwin saw the evolution of these tubes and tongues as the result of a race between flower and insect. If an insect could drink nectar without pressing its head against the orchid, it couldn't pass on its pollen. Natural selection would thus favor orchids with longer tubes. At the same time, an insect with a tongue that was too short for the tube wouldn't be able to drink all the nectar.

In some cases, this race between orchid and insect might drive each partner to absurd extremes. Darwin once received an orchid from Madagascar, called Angraeceum sequipedale, with a whip-shaped nectary over eleven inches long, with a drop of nectar tucked away at its very base. Only an animal with a suitably long tongue could drink it. Darwin predicted that somewhere in Madagascar there must live just such an insect.

The orchid's pollen, he declared, "would not be withdrawn until some huge moth, with a wonderfully long proboscis, tried to drain the last drop."

When Darwin died in 1882, the Madagascar orchid was still without a partner. But in 1903 entomologists discovered an extraordinary Madagascar hawkmoth. Normally its proboscis remained curled up like a watch spring. But when it approached orchids, it pumped fluid into the proboscis to straighten it out like a party balloon, and then insert it into the flower, as carefully as a tailor threads a needle's eye.

Scientists have found many other orchids and other flowers with an equally intimate relationship with their pollinators. Steven Johnson, a South African biologist, has documented lots of them in his part of the world, as he descirbed in an excellent article this spring in Natural History.

Now, in the August issue of the American Journal of Botany, Johnson and his colleagues have published a paper about a new orchid, shown in this picture. Disa nivea is a rare orchid found only in a few places in South Africa, and until Johnson came to study it, no one knew how it was pollinated. After a lot of patient orchid-watching, he and his colleagues discovered that it is visited exclusively by the fly shown in the picture. Its proboscis is well-matched to the length of the orchid, and the orchid grows pollen in just the right place so that they get stuck to the fly. You can see them in this picture--the two dangling yellow packets on the fly's snout.

There's just one catch: when the fly manages to get its proboscis all the way down to the bottom of the orchid's nectary, it finds no nectar.

To explain this deceit, Johnson and his colleagues observe that the orchids are always found intermingled with a similar-looking plant related to foxgloves. These plants are also pollinated by the same fly, but unlike the orchid, they reward visiting flies with nectar. Johnson and his colleagues argue that the orchid has evolved to mimic the rewarding flower, luring the flies with the same cues but deceiving them in the end.

To test this hypothesis, the scientists looked at five populations of the rewarding flower, measuring their dimensions. They found that from one population to another, the orchids mimic their local models. In some places, the rewarding flower is twice as long as in other places; the same goes for the orchid. Where the rewarding flowers are wide, so are the orchids; where they are narrow, the orchids are as well. These patterns are evidence that the evolution of this deceit is not a thing in the past, but an ongoing process.

Darwin would have not believed that such a deceitful plant could exist. Botantists had reported nectarless orchids as early as 1798, but Darwin thought they had to be wrong. Insects were too smart to be fooled for long. They would learn how to recognize a deceitful plant and avoid it, and the deceivers would become extinct. That turns out to be quite wrong. Over 8,000 species of orchids are believed to practice deceit. Most, like Disea nivea, mimic a food-supplying plant in their shape and odor. Others lure flies with growths that look and smell like feces. Others produce sex pheromones to lure male insects and sometimes even produce shapes that look and feel like female insects--so much so that the males try to mate with them. (More on wasp-on-orchid kinkiness here.) Orchids can in fact outfox insects, but only by continually reshaping their deceptions. Scientists suspect that the main benefit of deceit is that insects tend to fly far away after getting fooled. As a result, tend to fertilize more distant orchids, which gives the flowers a healthy supply of genetic variation.

It's fascinating to compare the story of Disea nivea to Angraeceum sequipedale. In one case, Darwin was right, and in the other he was wrong--at least in the details. His rough ideas about coevolution have developed over nearly 150 years into a huge body of knowledge about how partners shape one another over time. It just turns out that sometimes coevolution can push life in directions he couldn't imagine.

(Note: I adapted parts of the historical material in this post from my book Evolution.)

Update, Sunday 2 pm: For some reason the comments aren't going through for this post. We'll try to fix the bug today.

Update, Monday 11 am: Okay, comments are working again.