Counting the number of herbivorous insect species that scientists have identified reveals a remarkable possibility: insects that feed on living plants may be the most species-rich group of organisms ever to have evolved in the history of life on the planet. This includes extinct lineages such as the Palaeodictyoptera — among the first herbivorous insects to have evolved. Only unnamed microbes might be more diverse (although this hypothesis is controversial). Parents in some future colony on Mars could someday read to their children the bedtime story The Very Hungry Caterpillar by Eric Carle and with that story illustrate the essential nature of the two most diverse life forms on our planet: herbivorous insects and flowering plants.
In fact, together, herbivorous insect species and flowering plant species account for well over half of all species on the planet. By focusing on the mechanisms driving the diversification of these two lineages — these two branches of the tree of evolutionary life — we can begin to address the broader question, “Why are there so many species of life on earth?”
This big question bedeviled biologists well before Charles Darwin and Alfred Russell Wallace co-discovered the theory of evolution by natural selection in the late 1850s, and it remains multifaceted, nuanced, and nebulous. Nevertheless, although far from having all of the answers, the field of evolutionary biology has made considerable progress toward them.
Darwin hinted at the idea that species interactions are a crucible for the evolution of new species in one of the most beautiful passages in On the Origin of Species:
It is interesting to contemplate an entangled bank, clothed with many plants of many kinds, with birds singing on the bushes, with various insects flitting about, and with worms crawling through the damp earth, and to reflect that these elaborately constructed forms, so different from each other, and dependent on each other in so complex a manner, have all been produced by laws acting around us…Thus, from the war of nature, from famine and death, the most exalted object which we are capable of conceiving, namely, the production of the higher animals, directly follows. There is grandeur in this view of life, with its several powers, having been originally breathed into a few forms or into one; and that, whilst this planet has gone cycling on according to the fixed law of gravity, from so simple a beginning endless forms most beautiful and most wonderful have been, and are being, evolved.
What did Darwin mean by the “war of nature”? This phrase encapsulates both competition within species and antagonisms between species, such as predator-prey or plant-herbivore interactions. The war of nature then may help us understand why there are so many species of life on earth.
Evolving Apart, Together
When I was young, my family moved from Duluth, Minnesota, on the westernmost point of Lake Superior, deep into the boreal forest of St. Louis County. We lived alongside a large peat bog stretching for miles, from Sax and Zim to the Finnish town of Toivola, only twenty-five minutes from Hibbing, where Bob Dylan was raised. In this bog and the surrounding woodlands could be found great gray owls, timber wolves, tamarack trees, and, from time to time, forest tent caterpillars, which during outbreak years were so abundant that roadways became slick with their remains.
The summer we arrived — mercifully, we did not make the move during winter — I found myself sitting one afternoon near a patch of common milkweed, which was growing alongside an old railroad grade. I had heard that one should be able to find the caterpillars, or larvae, of monarch butterflies on the foliage of milkweed plants. Indeed, adult monarch butterflies were nectaring at the milkweed flowers, and I watched as the female butterflies alighted on the plants, curled their abdomens, and deposited small, sculpted eggs under the leaves. And sure enough, I also found the caterpillars, small and large, chewing on leaves in plain sight.
Why were the larvae so brilliantly colored? Wouldn’t birds be able to find them and eat them? These butterflies seemed out of place in the middle of the North American continent, to which they migrated every year with other gaudy creatures such as the scarlet tanagers, which nested in the tall sugar maples just a few hundred feet from where I sat. I knew that the tanager males, and not the females, were a brilliant red. Might this be a clue as to why monarchs are so brilliantly colored? Tanager males evolved to become red through the process of sexual selection because female tanagers are attracted to the color red. But this couldn’t be the case for monarch butterflies, since both male and female monarchs (as well as their larvae) are brightly colored. So what was the explanation?
Darwin and Wallace exchanged letters on this very topic in 1867. They came up with the idea that warning coloration, in the case of caterpillars that had fed on toxic host plants, could have evolved as a means to signal toxicity to predators. In 1890, the British biologist Edward Poulton coined the term “aposematism” for this phenomenon: Venomous or toxic animals the world over, including some caterpillars, stingrays, bees, coral snakes, and skunks, have evolved warning coloration as a way to signal their noxious potential. This explained the brilliant color of monarchs: Over time, they had sequestered chemical compounds from the milkweed plant that made them noxious to birds. A photo by monarch expert Lincoln Brower, published in Scientific American in 1969, shows a blue jay vomiting after consuming a butterfly, illustrating how this warning color works. Blue jays and other birds learn to avoid such brightly colored insects, which make them ill after ingestion.
I pulled off a few leaves of the milkweed and, true to its common name, out seeped a white, sticky latex. Within the latex are molecules called cardiac glycosides, forms of which occur in all milkweed species. These molecules diminish the functioning of a pump-like mechanism in animals’ cells that moves sodium and potassium. This has a strong effect on the cardiac muscles of animals, because cardiac cells rely on carefully regulated salt concentrations in order to contract.
Milkweed is not the only plant that has evolved to develop the ability to synthesize cardiac glycosides. The foxglove plant did as well, even though the two plants are from different evolutionary lineages with independent histories. So both plants contain a highly effective insecticide. However, monarchs — as well as several other species, including true bugs and leaf beetles — are resistant to the toxic effects of the glycosides, but they feed only on the milkweed plants.
As we will see, by looking more closely at this phenomenon — how diverse plants containing toxins interact differently with diverse species of herbivores — we can begin to understand how this might have facilitated the diversification of species.
In 1959, the zoologist Gottfried Frankel published a paper in the journal Science in which he proposed that the myriad so-called secondary compounds made by plants, such as the milkweed’s sticky latex, serve as defensive molecules. These compounds are called “secondary” because they serve no apparent function to the growth and reproduction of the plant. By contrast, primary compounds, such as amino acids, contribute directly to growth and metabolism; without them, plants could not live.
A few years later, in 1964, Paul Ehrlich and Peter Raven proposed that patterns of variation in secondary compounds observed across the diversity of plant life are linked to patterns of specialization among insects, such as caterpillars, specifically in how they feed on plants. Ehrlich and Raven noted that particular lineages of plants that produced distinct toxins, such as mustard-oil-bearing plants in the broccoli family, were often attacked by caterpillars found only on those host plants. It appeared that the specialization of secondary compounds in plants, which serve the purpose of self-defense, helped to produce specialization of insect species able to break through the defense, and vice versa, and that this was a common pattern across the diversity of plants and butterflies.
Why had this pattern evolved? Ehrlich and Raven suggested that alternating bouts of natural selection between the interacting plants and butterflies, a process called coevolution, could have driven the diversification of each lineage over time.
Imagine a scenario in which a plant species is attacked by herbivores but suddenly evolves a new secondary compound through a mutation in an existing metabolic pathway. Immediately, this plant is protected from attack by the very herbivores that adapted to the toxins previously present. This plant species is then likely to be a better competitor, losing less leaf tissue to herbivores than plants that do not make the new toxin. As a result, the plant that produces the new toxin is more fit than the others, producing more seeds, and, for a time, winning the war of nature. These plants eventually expand their range, invade new habitats, and differentiate into new species (that is, they “speciate”). Meanwhile, a mutation arises in a detoxification pathway in the herbivorous insects that allows them to overcome these newly evolved plant defenses. These insects are then able to escape not only competition for food with other herbivore species but also predation by other insects, such as wasps, who still look for them on their ancestral host plants. These insects track the range of the new host plant species and, like these plants, diverge and speciate.
This escape-and-radiate hypothesis helps to explain how coevolutionary dynamics could generate the diversity of plants and insects living on our planet. It also helps to explain how and why the diversity of secondary compounds evolved in plants, which humans exploit for dietary, medical, and recreational use.
May Berenbaum, a professor of entomology at the University of Illinois at Urbana-Champaign, has been studying interactions between parsnip plants and webworm caterpillars that feed on parsnips. Parsnip plants are in a family that includes dill, poison hemlock, and anise. Some plants in this family contain toxins called furanocoumarins, which can cause a skin condition in humans that is activated in the presence of ultraviolet radiation (phytodermatitis). These toxic compounds are also present in plants in the citrus family and are the reason some people should not drink grapefruit juice with certain medications.
Parsnips and webworms are not native to North America, but arrived with European settlers. This has allowed for a natural experiment to unfold that helps us to determine if the patterns observed across the diversity of plants and herbivorous insects developed at the population-level as well. Dr. Berenbaum found that different populations of parsnip were evolving new combinations of furanocoumarins and that these plants could sometimes escape attack by the webworms that relied on them as host plants. In turn, some webworms became locally adapted to the particular blends of furanocoumarins produced by their host plants. This showed that a coevolutionary arms race could be observed even within a human lifetime and that, indeed, plant-herbivore chemical coevolution might help us to understand how there have come to be so many species of plants and insects.
Clues in the Genetic Record
But although chemical coevolution may tell us a great deal about how specialization evolves in herbivorous insects, it doesn’t tell us how herbivory — the eating of plants — evolved in the first place. Understanding this question is an important piece of the puzzle in addressing the question “Why are there so many herbivore species?” Paradoxically, of the thirty insect orders that are currently living, from mayflies to beetles, only nine of these orders contain herbivorous species. This suggests that the switch to feeding on living plants might be challenging for nascent herbivores. In addition to overcoming plant toxins, these insects also face new, aerial habitats and the low nutritional quality of living plant tissues, compared to feeding on other animals. Once these barriers are overcome, however, herbivorous insect lineages produce new species at about twice the rate of non-herbivorous insect lineages. Thus, the evolution of herbivory may open a new adaptive landscape into which new species are born, as these insects adapt and radiate out into their new habitats.
In my own research laboratory at the University of California, Berkeley, we study how a fly species called Scaptomyza flava, closely related to the venerable genetic model fruit fly Drosophila melanogaster, has evolved to eat only the leaves of mustard plants. The ancestors of the leaf-feeding fly probably ate yeast, which often grows on plants, and this allows us to understand from the perspective of genomics how, potentially, evolutionary transitions to herbivory proceed.
This idea was taken up by a Ph.D. student in my group, Ben Goldman-Huertas, who wanted to understand how S. flava left behind a preference for yeast in favor of living leaves. Ben and I, along with our collaborators, found that S. flava had lost several genes that encode olfactory receptors known to be important for finding yeast in the fruit fly and many other microbe-feeding relatives. These receptor proteins are expressed in sensory neurons within small hairs on the third antennal segment — what is, by analogy, the nose of the fly. In some cases, we could find remnants of the genes in the S. flava genome, but they had accumulated random mutations and were no longer functional. Instead, the genes persist as genomic fossils, allowing us a glimpse of how the preference for yeast might have been lost. The S. flava flies are no longer attracted to yeast, and their antennae only weakly respond to its odors.
S. flava flies feed on plants with glucosinolates — the precursors to mustard oils (think wasabi) — and use the same metabolic pathway as humans to detoxify them. (Although S. flava shows evidence for adaptation at the molecular level in the enzymes that catalyze the detoxification of glucosinolates.) Mustard oils are potent insecticides that have evolved only in a certain order of flowering plants known as the Brassicales, as well as another, unrelated genus called Drypetes. Evolutionarily, the genes that help synthesize the glucosinolates are derived from the genes that helped synthesize precursor compounds called cyanogenic glucosides, which produce toxic hydrogen cyanide in plants. Comparative genomics across plants reveals that the evolutionary ancestors of glucosinolates are chemically similar to the cyanogenic glucosides. So it seems likely that the precursor to the Brassicales, which evolved to develop glucosinolates, lived in an environment lacking predators for a time before radiating and then, ultimately, getting colonized by herbivores such as S. flava.
Such findings allow us to see how compounds such as caffeine, for instance, evolved to deter feeding by herbivores or attack by other organisms, such as microbes. Although caffeine stimulates human nervous systems, it did not evolve in order to help us get through the day, but rather to kill smaller animals, thus protecting the plants from their predators. (High concentrations of caffeine are deadly even to humans, although you would need to drink a lot of coffee in one sitting for it to kill you.) When sprayed on plant leaves, even relatively diluted caffeine makes for an excellent insecticide. (Don’t try this at home—if only because of the cost.) The hypothesis that caffeine evolved to kill herbivores, however obvious to us today, was the subject of a breakthrough report published in Science in 1984.
Chemical coevolution between plants and herbivorous insects has produced not only a vast range of drugs that have been exploited by humans for millennia but also a profusion of species. This, then, may help us understand how plants and the insects that eat them — which constitute most species of life on earth — evolved. Modern genomics, coupled with classic evolutionary theory and natural history observations allow us to see how this process unfolds.
In my own laboratory, we are working on a new project with collaborators that aims to understand how the sodium-potassium pumps of milkweed-feeding and foxglove-feeding insects have evolved to resist cardiac glycosides. We are engineering fruit flies that have the same mutations in the sodium potassium pump as those found in monarch butterflies. This will allow us to understand the molecular basis of adaptation in extreme detail. From that first glimpse I had as a teenager into the chemical coevolution between tropical monarchs and common milkweeds in a field in northern Minnesota, my own intellectual life has come full circle.
- Does Darwin’s phrase “war of nature” seem to capture how we got from few species to so many?
- How does coevolution differ from straightforward evolution?
- How does coevolution help to explain the diversification of species?
- How might one test for coevolution in natural systems?
- What role do secondary compounds in plants play in coevolution?
- Might there be another answer to the question, “Why are there so many species?”
One major theme emerging from the questions raised in our discussion was the concept of diversity itself. We asked “Why are there are so many species?” But one might turn the question on its head and ask “Why aren’t there more?” Indeed, there are ecological and evolutionary hypotheses that address this question. We traced some of the major ideas in ecology that were developed in the twentieth century to explain patterns of biodiversity by Hutchinson from a niche-based point of view and by McArthur and Wilson from an island-biogeography point of view. We also touched on evolutionary ideas of speciation and extinction, specifically, that the tempo of the production and loss of lineages through time to yield a net diversification rate plays a critical role in determining past or present diversity in any group or any given habitat on the planet. Thus, historical factors and contemporary factors both play a role in determining why there are so many (or so few) species. Herbivores are a very successful group with an excellent record, robust molecular phylogenies (evolutionary trees) and abundant ecological data. Hence they serve as a good model for understanding the forces shaping current patterns of diversity.
The second theme that emerged was the idea of co-evolution — the meaning of the term and its history. Co-evolution does not imply that organisms adapting via co-evolution are somehow not subject to the same evolutionary constraints as those adapting to the abiotic environment. However, in a co-evolutionary scenario, the agents of natural selection are the traits of other organisms, which, themselves, have reciprocal effects on each species. Darwin alluded to co-evolution as an idea, particularly in relation to traits in the orchid plants he studied as well as their pollinators. He used the principle famously to predict a long-tongued moth that would pollinate a Malagasy orchid.
New Big Questions:
- Are we in a better position today to predict why a particular number of species persists on Earth than G. Evelyn Hutchinson originally thought?
- Is co-evolution testable at a grand scale, or is it simply a useful (or not) concept?
- Which are more important in shaping local, regional and global biodiversity: ecological or evolutionary forces?
- Is evolution predictable? Could we use plant-herbivore interactions as models to test this hypothesis?