Ecological Archives E096-233-A1
Connor R. Fitzpatrick, Anurag A. Agrawal, Nathan Basiliko, Amy P. Hastings, Marney E. Isaac, Michael Preston, and Marc T. J. Johnson. 2015. The importance of plant genotype and contemporary evolution for terrestrial ecosystem processes. Ecology 96:2632–2642. http://dx.doi.org/10.1890/14-2333.1
Appendix A. Supplementary methods of plant genotype and tissue collection, insecticide control experiments, and variance explained calculations.
A.1 Plant genotype and tissue collections
The 18 Oenothera biennis genotypes used in the experimental assays were selected to represent unique microsattelite genotypes, and the breadth of phenotypic diversity from a total of 40 genotypes grown in a common garden in 2006 (see Johnson et al. 2009). The 40 genotypes were originally collected from early successional sites, representative habitats of O. biennis. All collection sites are in Tompkins County, NY (one genotype per site, sites separated by an average of 12 km). The multi-year field experiment took place in Tompkins County, NY, separated by at least 2 km from original collection sites. The genotypic richness of the original populations (18 genotypes) is high but within the plausible range of genotypic diversity within natural populations (Johnson 2011). For our ecosystem assays we used seeds collected from individuals of the same genotypes grown in a common garden. We then grew plants from seed in a new common garden at the University of Toronto Mississauga greenhouse and used tissue from these individuals for our field and lab based assays.
A.2 Insecticide control experiments
We performed experiments to test for direct effects of insecticide on our measures of ecosystem function. To simulate the experimental conditions, we grew a total of 20 O. biennis plants in soil collected from an adjacent but unsprayed location at the field site in Ithaca. The plants were grown in 1L pots in an environmental chamber (12 h photoperiod and 24° C), under well-watered conditions. We subjected half of these plants to the same insecticide treatment as was used in the field experiment (biweekly insecticide spraying). After 4 months we harvested the soil from the sprayed and unsprayed pots and performed a lab-based leaf decomposition experiment and a seedling performance experiment using this harvested soil. Using the same protocols as outlined in the previous sections, we assessed the effect of sprayed and unsprayed soil on leaf decomposition using litter from one O. biennis genotype (n = 20) and the seedling performance of two O. biennis genotypes (n = 40). We selected these particular genotypes based on their strong interaction with soil collected from sprayed and unsprayed plots in the field experiment. We found no significant effect of insecticide application on either seedling performance (F1,38 = 0.09, P = 0.77), or the rate of litter decomposition (t17= -1.55, P = 0.14).
A.3 Variance explained
We used a model with only random effects fit by restricted maximum likelihood (REML) for each ecosystem response to estimate the proportion of total variation explained by plant genotype, herbivory, plot, and residual error. Summing the variance components and dividing the variance associated with each factor by the sum gives the proportion of total variance explained, which is equivalent to the coefficient of determination (i.e., R²). For continuous factors (genotypic composition axes 1 and 2, and Eudlidean distance), we used the lm function in R to perform a simple linear model where the response variable was the residuals from our final mixed-effects model with the continuous factor absent, and the predictor variable was that continuous factor. The R² of this linear model is the variance explained by genotypic composition after accounting for the variance explained by all other factors. We repeated this analysis for genotypic composition 1 and 2, and Euclidean distance for each ecosystem response. We present these results in Appendix C: Fig. C1.
Literature cited
Johnson, M. 2011. The contribution of evening primrose (Oenothera biennis) to a modern synthesis of evolutionary ecology. Population Ecology 53:9–21.
Johnson, M. T. J., A. A. Agrawal, J. L. Maron, and J. P. Salminen. 2009a. Heritability, covariation and natural selection on 24 traits of common evening primrose (Oenothera biennis) from a field experiment. Journal of Evolutionary Biology 22:1295–1307.