Zebra mussels (Dreissena polymorpha) are commonly employed as contaminant biomonitors in Europe and North America. Accumulation of hydrophobic, organochlorine contaminants into an organism depends, in part, on the organism's lipid content. In this study, we assess the influence of temporal variation in lipid content and reproductive activity on contaminant dynamics in zebra mussels collected from the western basin of Lake Erie. Mussels spawned twice during 1995, first in June and again in August. Lipid content of mussels was highest following the first spawning event and declined thereafter. Lipid-adjusted concentrations of moderately hydrophobic (mid-K o w ) compounds were much higher than those of either low or high hydrophobicity. Principal component analysis (PCA) of seasonal variation of concentrations of nine compounds revealed two primary axes, the first determined by mid- and high-K o w compounds (PCBs 52, 101, 118, 153, 138 and 180), the latter by a low K o w compound (hexachlorobenzene). A second principal component model separated physiological, environmental and temporal factors. Variation in mid- and high-K o w compounds corresponded with precipitation and reproductive status of mussels. By contrast, variation in hexachlorobenzene concentration was associated with lipid content and reproductive cycle. Concentrations of all nine compounds did not differ significantly between gravid and laboratory-spawned mussels, indicating that the effect of reproduction may be decoupled from the actual spawning event. Concentrations did not differ between spent male and female mussels. Examination of lipid composition revealed that neutral lipids comprise only 39% of total lipids in zebra mussels. Concentrations of PCBs with log K o w >5.71 were more highly correlated with neutral than with total lipid content, indicating that concentration normalization on a neutral lipid basis may be more appropriate than adjustment based on total lipid. Results from this study indicate that reproductive status and lipid content affect contaminant dynamics in zebra mussels and that these factors require consideration when mussels are deployed as biomonitors.