Plant chemistry and evolutionary history help to understand tropical biodiversity
Researchers from several disciplines, many of whom are based at the Institute of Entomology, collaborate on projects combining evolutionary history and plant chemistry to understand and predict the occurrence of herbivorous across rainforest plants. Their recent work has been published in the prestigious journal, Proceedings of the Royal Society B.
Figure - Co-phylogeny between (a) geometrid and (b) pyraloid caterpillars. Heat maps superimposed on the geometrid host phylogeny show observed abundance (left column) and the predicted probability of occurrence (right column).
Our team studies contemporary patterns in tropical diversity, and links them to evolutionary history and plant defensive traits. By studying these patterns, we hope to further understand the processes responsible for the incredible diversity of tropical insects. Here, we studied leaf chewing caterpillars in a New Guinean rainforest. These herbivorous insects are both species rich and ecologically important as consumers of plant biomass, and some of them are important pests. Understanding how caterpillars have diversified can go some way to help us understanding why there are so many species of insects and plants on earth. Some theories suggest that the interactions between insects and plants themselves have triggered diversification of one or both partners, while others suggest that there are simply lots of insects because there are lots of plants. Here we used leaf chemistry and the evolutionary relationships between both caterpillars and the plants that they eat to try and distinguish between these two scenarios. We did this by comparing how useful plant evolutionary history, insect evolutionary history and plant chemistry were for predicting what caterpillars eat. Our results suggest that one family of caterpillar is actually well adapted to chemically rich plants regardless of evolutionary relationships. Members of a second family of caterpillars remain more faithful to particular plant groups, suggesting that co-adaptation with plants may have occurred. We also use this information to make predictions of what caterpillars eat and successfully guess the majority of individual level interactions, suggesting that we can start to understand some of the complexities in tropical rainforest food webs using this approach. Accurate predictions include several important forestry pests, suggesting that our approach can be used to predict the interactions of invading species and emerging pests.
Segar, S.T., Volf, M., Isua, B., Sisol, M., Redmond, C.M., Rosati, M.E., Gewa, B., Molem, K., Dahl, C., Holloway, J.D., Basset, Y., Miller, S.E., Weiblen, G.D., Salminen, J-P., Novotny, V. 2017 Varyingly hungry caterpillars: predictive models and foliar chemistry suggest how to eat a rainforest. Proc. R. Soc. B 20171803. http://dx.doi.org/10.1098/rspb.2017.1803