THE ECOLOGY OF HOST PLANT-INSECT HERBIVORE INTERACTIONS IN THE FOSSIL RECORD FROM BIPARTITE NETWORKS

Plant–insect associations have been a significant component of terrestrial ecology for over 400 million years. Exploring these interactions in the fossil record through new perspectives provides a window into understanding the evolutionary and ecological forces that shaped these interactions. For the past several decades, researchers have documented, described, and categorized fossilized evidence of these interactions. Drawing on powerful tools from network science and neo-ecological studies, we propose a novel bipartite network representation of fossilized plants and herbivore-induced leaf damage to understand late Paleozoic plant–insect interactions at the system level.

A bipartite network is a particular type of complex network that has nodes divided into two sets. One set, X (for example, plant hosts) and the other set Y (for example, insect damage types that are also known as DTs) are joined by links occurring between these two sets but not within each set. Frequency data from several herbivory indices, such as damage type richness, surface area herbivorized, feeding event occurrences, and generalist-to-specialist herbivory values, can provide information from which links are built in these bipartite networks. These links define network properties using interaction strengths that express various aspects of herbivory richness and intensity. In this work, we focused our analysis on four Paleozoic plant biotas (from latest Pennsylvanian to early middle Permian) and their associated herbivore damage from north-central Texas, namely Williamson Drive, Mitchell Creek Flats, Colwell Creek Pond and South Ash Pasture.

We further explored how 16 network-based indices of centrality, co-occurrence, nestedness, robustness, and specialization shed light on how insect herbivore component communities were structured in the deep past. Overall, we found clustering of network properties based on the localities rather than by the type of insect damage or identity of the plant clade. Although we limit our analysis to only four sites in the current work, the methods used here can be applied to any part of the fossil record where preservation is sufficiently good to preserve leaves and their associated insect damage.