Wednesday, August 6, 2008 - 4:00 PM

COS 73-8: Germination responses of an annual plant in native and non-native ranges: Environmental risk selects for bet-hedging

José L. Hierro1, Özkan Eren2, Liana Khetsuriani3, Alecu Diaconu4, Katalin Török5, Daniel Montesinos6, Krikor Andonian7, David Kikodze3, Levan Janoian8, Diego Villarreal9, Eugenia Estanga-Mollica9, and Ragan M. Callaway10. (1) CONICET and Universidad Nacional de La Pampa, (2) Adnan Menderes Üniversitesi, Turkey, (3) Institute of Botany of the Georgian Academy of Sciences, Georgia, (4) Institute of Biological Research, Romania, (5) Institute of Ecology and Botany of HAS, Hungary, (6) Centro de Investigaciones sobre Desertificación, Espańa, (7) University of California, Santa Cruz, (8) American University of Armenia, Armenia, (9) Universidad Nacional de La Pampa, Argentina, (10) The University of Montana

Background/Question/Methods

Human-assisted introductions provide powerful systems for exploring classic ecological and evolutionary concepts such as genetic differentiation and local adaptation in populations. A number of species have been successfully introduced to several different regions, which often vary in biotic and abiotic conditions. The introduction of conspecifics into areas with contrasting selective forces allows examination of one of the most fascinating questions about local differentiation: does divergence among populations in the non-native range relate to divergence among populations experiencing similarly contrasting selection agents in the native range? We addressed this question by conducting a series of common garden experiments where we investigated germination responses of populations growing in environments with summer drought versus summer rain conditions within both the native and non-native range of the annual ruderal Centaurea solstitialis L. (yellow starthistle, Asteraceae). Additionally, we correlated germination proportions with the total amount and variation of annual and seasonal precipitation experienced by native and non-native populations. 

Results/Conclusions When we based comparisons on general climatic patterns (i.e., summer drought versus summer rain), all native populations, irrespective of the climate in which they occurred, and non-native populations occurring in regions with dry summers displayed similarly high germination proportions and rates, whereas genotypes from the non-native region with predominantly summer rain exhibited lower germination fractions and rates. In contrast, germination proportions of populations in both native and non-native ranges were strongly correlated with variation in precipitation during the winter, which is the season that immediately follows germination of C. solstitialis. In this correlation, as predicted by bet-hedging theory, germination fractions were lower in native and non-native populations with greater inter-annual variation in winter precipitation. These findings together with information on molecular genetics and the introduction history of the species suggest that natural selection in native and non-native ranges has favored large annual germination fractions in environments with low variation in winter precipitation and small germination fractions (i.e., large proportions of dormant seeds) in environments with high variation in winter precipitation.