No simple sum: seasonal variation in tannin phenotypes and leaf-miners in hybrid oaks Christopher T. Yarnes 1,2,4 , William J. Boecklen 1,2 and Juha-Pekka Salminen 3 1 Laboratory for Ecological Chemistry, Institute for Natural Resource Analysis and Management, Las Cruces, New Mexico 88003,USA 2 Department of Biology, New Mexico State University, Las Cruces, New Mexico 88003, USA 3 Laboratory of Organic Chemistry and ChemicalBiology, Department of Chemistry, Universityof Turku, FI-20014 Turku, Finland 4 Current address: Emory & Henry College, Biology Department, Box 947, Emory, VA 24327, USA Abstract. Parent-hybrid asymmetries in the biosynthesis of defense chemistry are believed to affect the distribu- tion of herbivorous insects within plant hybrid zones. While tannins are often produced as complex, dynamic mixtures, the ecological effects of biological and ontoge- netic variation in tannin metabolism within hybrid zones are poorly understood. Here, we examine correlations between the seasonal variation of absolute and relative concentrations of individual ellagitannins and total pro- anthocyanidinsinthe Quercusgambelii  Q.grisea hybrid complex and a community of leaf-mining Lepidoptera (Gracillariidae, Heliozelidae, Nepticulidae, Tischerii- dae), and examine the interactive effects between tannins on leaf-miner abundance. We found strong seasonal variation in oak tannin phenotypes, but two general phenotypes are characteristic of oaks of the Q.gambelii  Q. grisea complex, based upon biosynthetic variation in high molecular weight ellagitannins. Leaf-miner com- munity structure was significantly correlated with the dominant tannin phenotypes in the Q.gambelii  Q.gri- sea hybrid zone in mid-summer (July, August). Addi- tionally, leaf-miner density was significantly negatively correlated with the ratio of vescalagin-derivatives and total proanthocyanidins in August and September. Overall, hybridization affects biological and ontogenetic variation in ellagitannin and proanthocyanidin biosyn- thesis within the Q.gambelii  Q.grisea hybrid complex and this variation correlates with herbivore distribution and abundance. Future studies of phenolic chemistry in plant-herbivore interactions in hybrid zones should in- clude a more rigorous quantification of tannin pheno- types as complex, temporally dynamic mixtures. Keywords. Hybridization – Quercus – ellagitannin – leaf- miner – phenotype – defense Introduction Hybridization between plant species often produces natural variation in host-plant morphology and function, including plant defenses. Hybridization may affect the quantitative and qualitative distributions of tannins, pu- tative plant defenses, through hybrid zones (Orians and Fritz 1995, Orians 2000, Fritz et al. 2001, O)Reilly-Wap- stra et al. 2005). However, most studies have focused on summary measures of tannins (e.g. Schweitzer et al. 2004, Bailey et al. 2004) that may have limited applicability for ecological studies (Appel et al. 2001, Heil et al. 2002, Salminen 2003). A more accurate characterization of tannin phenotypes is needed to clarify the fine-scale spatial and temporal variation in tannins within hybrid zones and the potential effects for associated communi- ties of consumers. Whiletanninshavehistoricallybeentreatedasasetof compounds with homogeneous ecological function, this class of phenolics represents a diverse assemblage of biochemicals with context-dependent biological func- tions (see Zucker 1981). Further, individual tannins have been shown to exhibit wide-ranging effects for different herbivores (Ayres et al. 1997). Moreover, there is a dis- tinct lack of empirical evidence about how tannin phe- notypes – the absolute and relative concentrations of in- dividual tannins as a complex mixture – may affect her- bivoredistributionandabundanceinnaturalpopulations. However, Barbehenn and colleagues (2006b) have re- cently demonstrated that at high pH, similar to that of insect digestive tracts, the oxidative capacity of individual ellagitannins is dependent upon structural characters, particularly the number of hydroxyl groups, and is me- Correspondence to: Christopher T. Yarnes, e-mail: cyarnes@ehc.edu. Chemoecology 18: 39–51 (2008) 0937-7409/08/010039-13 #Birkhäuser Verlag, Basel, 2007 DOI 10.1007/s00049-007-0391-y CHEMOECOLOGY