COS 74-1
Seed predators, not herbivores, exert natural selection on goldenrod (Solidago altissima L.) in an urban archipelago

Wednesday, August 12, 2015: 8:00 AM
344, Baltimore Convention Center
Robert Bode, Natural Sciences, Saint Martin's University, Lacey, WA
Adam Skowron, Biology, Canisius College
Austin Gilbert, Biology, Canisius College
Background/Question/Methods

In an urban environment, plants exist in isolated patches that are effectively islands in a sea of concrete and asphalt. Herbivores and seed predators that feed on these plants must colonize the islands, and different species do so with differing degrees of success. Past studies have shown that more arthropods exist on plants in larger urban patches, as predicted by island biogeography theory. We expected to see that, since large patches would host more arthropod herbivores and seed predators, plants in these patches would suffer more damage by herbivores and lose more fitness to seed predators. To investigate this, we looked at the levels of foliar damage and seed predation in goldenrod (Solidago altissima L.) in an urban setting in the city of Buffalo, New York, U.S.A. Goldenrod is found in small to large patches in abandoned lots and parks throughout this city, and has a wide range of herbivores that vary in abundance, depending on patch size. We measured the number of herbivore attacks per leaf, percent tissue removal, and percent seed predation as functions of patch size.

Results/Conclusions

We found that herbivore pressure increased with increasing patch size, but was relatively low throughout the study area. Both the number of herbivore attacks and the percent tissue removal were higher in larger patches, but percent tissue removal was less than 10% of total tissue. Since goldenrod can have up to half of its leaf tissue removed before suffering any reduction in fitness, we propose that the damage seen is not sufficient to represent a force of natural selection. Seed predation also was higher in larger patches, and represented a direct fitness cost to the plants. We believe this implies that natural selection in an urban system is primarily by seed predators rather than by herbivores. This may also imply that, since goldenrod grows clonally, as a patch expands, the natural selection by seed predators will increase as a function of time.