Background/Question/Methods The level of secondary plant metabolites such as polyphenols is dependent on genetic, ontogenetic and environmental factors.  However, variations in chemical composition of polyphenolics with these factors have not been examined in detail.  Compositional variation can lead to different reactivities of polyphenolics, with galloyl-substituted subunits more active as biological oxidants and catechol-substituted subunits more active as protein binding agents.  Reactivity differences may lead to functional differences.  For example, herbivores may have different sensitivities to dietary oxidative stress vs. protein limitation.  The fate of leaf litter enriched in oxidatively active, galloyl-substituted phenolics may differ from that of litter enriched in catecholic polyphenols.    The composition of condensed tannins can be determined by chemical degradation followed by HPLC analysis.  We have used a large tissue collection of Populus (cottonwood) leaves as a model system for understanding compositional variation.  The material includes leaves from ten individuals of each of four genetic cross types, collected over the growing season from both juvenile and mature developmental zones.  Plants were grown in gardens with different environmental conditions. 
Results/Conclusions To date, we have established that catechin is the common terminal unit for these flavan-3-ol polymers from four cross types.  Both chain lengths and composition vary with genetic background.  Narrowleaf Populus condensed tannins are enriched in galloyl-substituted (prodephinidin) proanthocyanidins, while Fremont, with much lower overall tannin content, comprises catechol-type (procyanidin) tannins.  Narrowleaf tannins have longer chain lengths than Fremont, although improved methods for determining chain length are needed.  The F1 plants contain tannins with composition more similar to one parent (Fremont) but tannin chain length that is intermediate between the parents.  The backcross to Narrowleaf yields plants with tannins not significantly different from Narrowleaf in composition or chain length. 
The same techniques were employed to evaluate leaf tissue from mature Narrowleaf from three different gardens.  There are no significant differences in tannin composition or chain length for the three gardens, suggesting that genetic factors dominate over at least some environmental factors.  A study examining mature vs. juvenile tissue from both Fremont and Narrowleaf is in progress. 
Tannin composition is under genetic rather than environmental control.  The distinct composition of tannins in the cross types of Populus present a natural system for examining the biological activities of distinct types of condensed tannins.  This could lead to predictions of biological consequences of tannins based on structure-activity relationships, and a deeper understanding of tannins in evolution and in current day ecosystems.