Investigating Plant-Arthropod Interactions

Photo by Papa Pic via Flickr

"I study the association between a group of neotropical longicorn beetles and their host plants, a family of tropical trees that include the Brazil nut," she told us. (Neotropical refers to the tropical regions of the Americas, as opposed to those in Africa, the Indian subcontinent, and Asia. Longicorn beetles are sometimes called long-horned beetles, nicknamed for their very long antennae, or timber beetles, because the larvae tunnel into wood).

Her research is extremely complex, requiring knowledge of arthropods and plants and the interaction between the two groups of organisms. "I intentionally selected a project that had a lot of components because I thought it would be more stimulating. When you're working toward a doctorate, you spend a lot of time with your subject, so I looked for something that would offer me variety."

She certainly got it.

Amy's arthropods are members of the Cerambycidae, a family of beetles that come in many shapes and sizes. "I spent a whole semester in the entomology collection at the American Museum of Natural History, drawing my way through the cerambycids." If the best way to understand an organism is through close observation, many investigators believe the best way to begin is by drawing. "It's a huge family, with 35,000 described species, and I was just trying to get the lay of the land. By the time I did that I was hooked, because they are so spectacular," she said.

Next, she went out into the field. "I spent a year in Central French Guiana doing rearing experiments. I knew there was a group of cerambycids that laid their eggs only on trees belonging to the Brazil nut family (Lecythidaceae). Organisms that are picky like that are referred to as specialists. And I knew that this tree family was avoided by the generalists, insects that lay their eggs on many different plants. I wanted to find out how these beetles were using this resource, so I had to look more closely at the relationship between the beetles and the trees."

Amy designed her study to include a number of variables in the hope that at least some of her questions would be answered. "I worked with five different tree species. My beetles are wood-borers, so we had to provide them with freshly cut wood. I had the assistance of a professional tree climber who climbed up into 25 huge rain forest trees. The tallest one was about 47 meters (more than 150 feet) high. I wanted to see if different beetles prefer different-size branches, so in each tree I asked him to cut a skinny branch and a thick branch. He left a section of each branch up in the canopy (treetop) and brought down a section so I could see if there are different beetles at canopy level and at ground level. We did this two times: first in the dry season and then in the rainy season.

"I looked at five different tree species, two different branch diameters, two different strata (levels) of the forest, and two seasons." That is a lot of variables. "My question was: Were the beetles dividing up the resource? That is, did some prefer certain trees? Could some be found only in the canopy, while others could be found only on ground level?"

"I couldn't afford to hire the climber more than twice, so when he climbed up to do the rainy season cuts, he brought down the branches that had been left up in the canopy during the dry season. And then we rigged up pulleys so when the time came to collect branches after the rainy season, I just pulled them down," Amy told us.

Each branch was left in the forest for three and a half months, to allow the beetles time to lay their eggs. Then Amy took them back to where she was staying, put each branch in a plastic cage, and waited to see what emerged.

"Part of what I was trying to find out was whether there was host specificity, that is, how picky are these particular insects? I was also trying to identify plant compounds involved in the host selection process. What was it about the trees I was studying that attracted my beetles but not other beetles?"

Here are some of the things Amy found out: Not all of the tree species were equally popular. "I had two popular tree species, one that was kind of so-so, and two that were really unpopular." As it turns out, the least popular tree—Couratari stellata—has a truly foul odor.

"It was so stinky you could smell it before you saw it when you were walking through the forest. I found very few species of beetles and very few individuals of each species in the branches from that tree. A second tree species also had the odor, but it was not nearly as strong. A lot of species reproduced in that one, but there were still not many individuals of any species." Amy has a hypothesis to explain this finding. "I know my beetles are capable of reproducing in those trees because I did get a few, but I think they don't like the smell any more than I do, so they avoid it if they can."

One thing that surprised Amy was that during the rainy season more beetles reproduced at canopy level, where it is drier and windier than at ground level. "I'm not sure why," she said, "but I do have a hypothesis." She thinks that during the rainy season the air at ground level is so full of moisture that the volatile chemicals that attract the beetles are unable to travel very far. "I think the beetles just couldn't find the branches at ground level," she said.

Amy found answers to quite a few of her questions, but what she mostly came up with were more questions.

"I ended up building a lot of new questions on my original questions." We asked her if she found it frustrating to come up with more questions than answers. "I don't think it's frustrating," she said. "I think it's fun. I really like coming up with possible explanations for what I observe, and I like generating ideas. To me, the best part of discovery is following a trail of questions."

Can Beetles Smell? 

If Couratari stellatais unpopular with cerambycids, does that mean these beetles can smell? "Yes, indeed," said Amy Berkov. "Insects are incredibly sensitive to smell compared to humans. They find their appropriate host by smell, and they don't make very many mistakes."

But what do they smell with, since they don't have noses? "They smell with their antennae, which are highly sensitive sensory structures, especially for smell but also for touch. Volatile chemicals waft through the air, and the beetles pick them up with their antennae and follow the smell, just like you might follow the smell of freshly baked cookies. Once they actually land on a tree they make sure they made the right selection by tasting the bark with their mouthparts. Some insects, including butterflies, moths, and flies, have taste receptors on their tarsi so they can actually taste with their feet."

How do Cerambycidae and Lecythidaceaea interact? 

Plants give off volatile chemical compounds that can be sensed in the air by insects. A few cerambycid species are specialists that are attracted to the particular compounds coming from Lecythidaceae. "The adult beetles meet and mate right on the tree, and then the female lays her eggs there," Amy Berkov told us. "When the eggs hatch, the larvae tunnel down into the wood."

Some cerambycids lay their eggs on rotting wood; others select intact, living trees. The particular cerambycids Amy studied favor freshly cut wood. "They are among the first insects to attack freshly killed wood. They spend the next four to five months in the larval state, chewing the wood and creating a system of tunnels and galleries inside the branch. Then other insects and microorganisms enter into the wood and speed up the process of decomposition."

Are there really 30 million arthropod species?

Part of what inspired Amy's research was a theory put forth by Terry Erwin in the early 1980s. Erwin suggested that there might be 30 million species of arthropods, implying that for every one that has already been described there are 29 that are unknown, unnamed, and unclassified.

Erwin's statement was based on the assumptions that many arthropods are very picky about where they live and lay their eggs (this is called host specificity) and that there are twice as many arthropods that live only at treetop level as there are that live on the ground. It was a provocative statement that scientists have been debating ever since. But, as Amy pointed out, "We need people to go out and make wild speculations because it focuses research on areas that have been ignored. Erwin made everybody realize that we don't know much about the host specificity of tropical arthropods, and we don't know whether the canopy and ground faunas are distinct, and we don't have any idea how many species exist. In fact, there is almost no data that allow us even to begin to evaluate those figures."

So what is the answer?

"I wasn't trying to prove or disprove what Erwin said," Amy explained. "What I wanted to do was take a small piece—sort of like a snapshot or a single piece of a complex jigsaw puzzleÐand examine it closely."

Still, looking at the snapshot, Amy's experiments with cerambycids did not support some of the assumptions on which Erwin based his statement.

"Erwin guessed that about two-thirds of the arthropods would be found only at canopy level and one-third only at ground level," she said. In the case of the beetles she was studying, only 17 percent were found at canopy level and 28 percent at ground level. More than half (55 percent) could be found at both levels.

Erwin thought there were at least 29 undescribed arthropods for every one that has been described, but only about one-third of the beetles in Amy's study are not yet described. "That is still a lot," Amy said, "but it's not even close to Erwin's projection. Of course, cerambycids are very large beetles—the megafauna of the insect world. In most cases, the big, conspicuous, and dramatic-looking organisms are much better studied than ones that are tiny or for other reasons more easily overlooked. I might have gotten a much higher ratio of undescribed species if I had been studying fungus gnats or something else small and humble."