Friday, August 7, 2009: 9:20 AM
Grand Pavillion II, Hyatt
Todd Wojtowcz1, Luke A. Hanna1, Liza M. Holeski2, Louis J. Lamit3, Richard L. Lindroth4, Thomas G. Whitham5 and Catherine A. Gehring6, (1)Biology, Northern Arizona University, Flagstaff, AZ, (2)Department of Entomology, University of Wisconsin, Madison, WI, (3)Biological Sciences, Northern Arizona University, Flagstaff, AZ, (4)Entomology, University of Wisconsin, Madison, WI, (5)Department of Biological Sciences and Merriam Powell Center for Environmental Research, Northern Arizona University, Flagstaff, AZ, (6)Merriam Powell Center for Environmental Research, Northern Arizona University, Flagstaff, AZ
Background/Question/Methods Litter systems and the arthropods inhabiting them play critical roles in nutrient cycling and may be important in the initiation and maintenance of plant-soil feedbacks. Interspecific variation in litter chemistry can influence litter arthropod abundance. However, little is known about the sensitivity of these arthropods or litter layer chemistry to genetic gradients created by plant hybridization. Using a 15-year-old cottonwood common garden representing the Populus fremontii x P. angustifolia hybrid system, we investigated the influence of the Populus genetic gradient on litter layer chemistry and litter arthropod abundance. Specifically, we tested two hypotheses. First, decomposing litter beneath Populus species and their hybrids vary predictably in lignin and bound condensed tannin (BCT) content. Second, litter arthropod abundance would be sensitive to the Populus genetic gradient and variation in litter chemistry. We focused our study on the responses of two common detritivores, isopods and millipedes, and two abundant and important groups of microarthropods, mites and Collembola.
Results/Conclusions Lignin and BCT content of decomposing litter decreased along the genetic gradient from P. angustifolia to P. fremontii. Approximately 50% and 30% of the variation in decomposing litter lignin and BCT content, respectively, was explained by the genetic gradient that resulted from hybridization. Isopod abundance was not responsive to the Populus genetic gradient, but 17% of the variation in millipede abundance was explained by the genetic gradient, with decreasing abundance from P. angustifolia to P. fremontii. The genetic gradient did not influence Collembola abundance but explained 19% of the variation in mite abundance, which declined across the genetic gradient from P. angustifolia to P. fremontii. Using structural equation modeling (SEM), we determined that host plant genetics influenced millipede abundance through lignin content, but not BCT content, of decomposing litter. In contrast, SEM indicates the genetic influence on mite abundance was not filtered through litter lignin or BCT content, but instead resulted from some other unmeasured variable(s). These findings suggest that plant genetic gradients can impact the arthropods most intimately associated with nutrient cycling processes and that multiple mechanisms, including genetic variation in litter chemistry, contribute to these relationships.