A study was conducted over the course of a year to determine the induction of hepatic cytochrome P4501A (CYP1A) in three species of benthic fish collected from a contaminated site compared to fish sampled from a less-contaminated site. Juvenile fish were used to minimize effects of reproductive status and migration. CYP1A was determined by two catalytic assays [aryl hydrocarbon hydroxylase (AHH) and ethoxyresorufin O-deethylase (EROD)I and by an immunoassay (ELISA) utilizing polyclonal antibodies raised against purified CYP1A from cod. AHH activities were measured by a standard method (AHH(std)) and by two variations of the standard method. All three primary CYP1A measures (AHH(std), EROD, and ELISA) showed consistent between-site differences, indicating that induction of CYP1A can be a reliable biomarker of contaminant exposure in fish if appropriate biological variables are controlled for in field studies. Multiple ANOVA demonstrated that the AHH(std) and ELISA data showed less variability due to species or temporal differences, and less unexplained variability, compared to the data from the EROD assay or either variation of the AHH assay. For all measures, variability associated with site differences far outweighed species or temporal variability. Immunoassay, while less sensitive than the AHH(std) assay, is nonetheless recommended to be used in conjunction with catalytic assays because of the potential for samples to lose catalytic activity if not handled properly. The current results suggest that the lower noncontaminant-related variability of AHH(std) makes this CYP1A measure potentially more useful for monitoring programs in which analysis of trends is a primary goal.