2004 Denver Annual Meeting (November 7–10, 2004)

Paper No. 16
Presentation Time: 5:15 PM


SMITH, Dena, CU Museum of Natural History, Univ of Colorado, Campus Box 265, Boulder, CO 80309, dena@colorado.edu

There is a growing interest in examining patterns of insect feeding damage on fossil leaf assemblages. Exciting patterns have emerged from this work including the geologically early establishment of major plant-feeding guilds, the identification of long-lived specialized interactions and a positive correlation between insect damage levels and climate. In this study, I focus on leaf mining, which is a feeding strategy that tends to be a very specific type of feeding done by predominantly specialist insects. Several examples of long-lived associations in the fossil record have been determined from leaf mines and some authors have cited the loss of host-specific damage in the fossil record as an indicator of insect extinction. I examined a modern leaf litter assemblage from a lowland tropical wet forest to examine the potential for bias in leaf miner damage levels. Leaf litter samples were collected from five transects in a lowland tropical rainforest in Costa Rica. Approximately 2,500 leaves were collected from all of the transects and examined for leaf mining damage. When present, I noted whether the leaf mining damage was apparent on the top of the leaf only, the bottom of the leaf only, or on both sides of the leaf. Only 6% of the 2,500 leaves examined had evidence of leaf mining damage. Six percent is much higher than levels typically seen in fossil assemblages. Seventy-nine percent of the leaves with leaf mining damage, had mines that were only visible on the upper side of the leaves. Nine percent of the leaf-mined leaves had mines that were apparent on the bottom only and 12% of the leaf-mined leaves had evidence of mines on both sides of the leaves. Because leaf mining damage tends to be much more apparent on only one side of a leaf, there is a very low probability that true leaf-mining levels will be observed in a fossil assemblage. This likely explains the very low numbers of leaf mining damage typically found in fossil assemblages (ie. 1 leaf mine per 600 leaves examined in one Cenozoic leaf assemblage). Studying the potential for bias in the preservation of other types of host-specific feeding damage, like galling, is likely to yield similar results. Caution should be taken when trying to use the diversity of feeding guilds as proxies for insect diversity.