Scars on fossil leaves: An exploration of ecological patterns in plant–insect herbivore associations during the Age of Angiosperms

Abstract

Ancient plant–insect herbivore associations can be studied directly through observation of feeding damage scars on well-preserved leaf adpression fossils. Early work on insect herbivory was largely qualitative and descriptive. The establishment of the insect damage census protocol by Wilf and Labandeira in 1999 modernized the study of ancient insect herbivory, formalized the functional feeding group–damage type (DT) system, and allowed rigorous quantitative analyses. In this review, we first discuss how to recognize insect herbivore damage and conduct insect damage censuses, as well as the importance of collecting these data. We then present a compilation of all published insect damage census data from angiosperm-dominated ecosystems that included at least 300 fossil dicot leaves. These 66 datasets range from Late Cretaceous to Pleistocene in age and include the majority of continents. Damage richness, damage frequency, relative abundance distributions of functional feeding groups, and ecological network metrics were computed for each site. We investigated spatio-temporal variations and tested potential drivers using various statistical analyses. Reassuringly, herbivory metrics do not appear to be influenced by publication date or depositional setting. There is no linear correlation between herbivory and geologic age, and the few significant differences that occur among time bins document the importance of the K-Pg extinction event on insect herbivory. When sites are partitioned into latitudinal bins, the mid southern hemisphere (60°S to 23°27′S) stands out as having frequent and diverse damage. High latitudes tend to have low damage richness and frequency, although not all differences are significant. Mean annual temperature influences herbivory more than any potential abiotic or biotic driver explored in our analyses and has significant positive relationships with total, specialized, and galling DT richness at 300 leaves, as well as specialized damage frequency. We close our paper by discussing best practices and promising avenues for future insect herbivory research.