There are more tree species in a half-hectare of tropical forest in Ecuador than in all of temperate Europe, Asia, and North America combined. This raises the question of how large numbers of ecologically similar species manage to coexist despite intense competition for light, water, space, and nutrients. One popular idea argues that insects may give an advantage to rare plant species by attacking successful plants when and where they become abundant. Yet this hypothesis could only work if insects have highly specialized diets. This research project will test that hypothesis through a study of the diet of all insects that are found to eat tropical forest plants in the coffee family. To this end, several thousand beetles, caterpillars, grasshoppers, and other insects will be collected in the rainforest in Panama. In order to identify the plants that a particular insect has eaten, DNA will be taken from the insect stomachs and the plant from which it came identified using DNA signatures unique to each species of plant. Plants will also be studied for differences in their chemistry that may protect them from insect pests. Results will be used to examine whether plant chemistry influences insect feeding, and whether insects are specialized enough to maintain the plant diversity of tropical forests by eating the most abundant plant species. Illuminating the role of insects in maintaining the tree diversity of tropical rainforests will shed light on one of the persistent mysteries of earth's most diverse ecosystem.
The proposed research requires molecular lab work at the University of Michigan as well as fieldwork at facilities of the Smithsonian Tropical Research Institute in Panama, providing opportunity for interdisciplinary training of students. Laboratory work in molecular methods will include training of two graduate students in a University of Michigan program for non-traditional and underrepresented students. Furthermore, this research will contribute samples and molecular data on insect and plant specimens to public collections at the University of Michigan and the Smithsonian Tropical Research Institute.
There are more tree species in half of a square mile of tropical forest in Ecuador than in all of temperate Europe, Asia, and North America combined. This raises the question of how large numbers of ecologically similar species manage to coexist despite intense competition for light, water, space, and nutrients. One popular idea proposes that insects may give an advantage to rare plant species by attacking successful plants when and where they become abundant. Yet this mechanism could only work if insects have highly specialized diets. My research funded by this award has examined the diet specialization of the insects that are found to eat tropical forest plants in the genus Psychotria in the Coffee Family. To this end, I have collected several thousand beetles, caterpillars, grasshoppers, and other insects in the rainforest in Panama with the help of undergraduate students and Latin American assistants. In order to identify the plants that a particular insect has eaten, we isolated DNA from the insect’s stomach and identified the plant from which it came using "DNA barcodes" unique to each species of plant. My collaborators and I have also studied differences in plant chemistry that may protect them from insect pests and dictate the diet, and hence specialization, of the insects. Many of the botanical drugs that humans know and find so addicting—from the caffeine in coffee to the nicotine in tobacco—most likely evolved as insecticides to defend plants from their natural enemies. What is more, such drugs are very often alkaloids, nitrogen-containing compounds that are diverse and common in the Coffee Family in general and in Psychotria in particular. This group gets its name from the psychotropic, or mind-altering, effects of one species found in the Amazon Rainforest of South America. We used chemical analyses and statistical models to examine whether plant chemistry influences insect feeding, and whether insects are specialized enough to maintain the plant diversity of tropical forests by eating the most abundant plant species. What we found was that of all of the species of Psychotria that live in Barro Colorado Island in the Panama Canal, the species that actually occur near one another are less similar chemically than expected by chance. This suggests that insects not only disproportionately attack abundant plant species, but attack chemically similar species where they co-occur, thus favoring groupings of chemically distinct plants. These findings suggest that plant-enemy interactions hold the key to understanding both the origin and the maintenance of the extreme diversity observed in tropical forest plant communities. Psychotria is one of the most diverse plant genera on the planet, with over 2,000 species, and can exhibit extreme local diversity. Our results suggest that environmental tolerance, such as tolerance of seasonal drought, may be an important determinant of species' distributions, but does not explain the evolutionary diversification of the genus or the coexistence of large numbers of species at local scales. Rather, our results suggest that the evolution of novel defensive traits—especially the alkaloids from which Psychotria gets its name—allows species Psychotria to escape the enemies they would otherwise share with their neighbors, thereby facilitating the coexistence of close relatives in shared habitats over ecological time, and promoting the diversification of hundreds of species of Psychotria over evolutionary time. On a practical level, our results suggest that phylogeny, or evolutionary relatedness, may not predict chemical similarities within species-rich genera of tropical rainforest trees. Instead, scientists looking to identify chemical similarities in tropical trees would do well to use the linkages between plants elucidated by their common enemies: the insects and pathogens whose host ranges follow patterns of chemistry shared between the plants. This project has required both molecular lab work at the University of Michigan (UM) as well as fieldwork at facilities of the Smithsonian Tropical Research Institute (STRI) in Panama, providing ample opportunity for the training of undergraduate and other students. Fieldwork in Panama has involved the mentoring of a UM undergraduate student sponsored by the NSF Research Experience for Undergraduates program, as well as a Colombian student sponsored by STRI and two Panamanian paid interns. Laboratory work at UM has involved the training in molecular methods of two UM undergraduates, one of whom was sponsored by the ED-QUEST program for underrepresented minorities in science. In addition, fieldwork in Panama has given me the opportunity to present my research findings to tour guides who lead walking tours of Barro Colorado Island. The majority of the tours cater to Panamanian school children, and I often contributed to the tours myself, in Spanish, when we would cross paths in the forest. Prompting an intellectual interest among Panamanian children in the diverse and fascinating rainforest that lies in their backyard, so to speak, has proved a gratifying way of sharing my scientific activities with the broader community.
