There is growing interest in the assessment of polychlorinated dibenzofurans (PCDFs) and polychlorinated dibenzodioxins (PCDDs) in fish. Over 180 documents have been reviewed in the development of the data base for this review article which includes nearly two thousand values for 2,3,7,8-TCDD concentrations and toxicity equivalencies (TEs) in fish in the United States and international waterways. The data indicate that there are fish with quantifiable levels of dioxin world-wide. This ubiquity is also seen in fish taken from US waterways. Specifically, only one state in fifty had 2,3,7,8-TCDD levels which were all below method detection limit, BMDL. All 50 states had fish which resulted in TE values above the MDL. Based upon the data assembled for this review article, the presence of PCDFs/PCDDs in fish appears independent of their feeding habits and habitats. A more focused review of fish in waterways in the United States resulted in the same conclusions.There is poor correlation (r2 = 0.692) between the average TE values and the corresponding average 2,3,7,8-TCDD concentrations. This is not unexpected given the multitude of natural and man-made sources of PCDFs/PCDDs in the environment, the different ratios of congeners present in different sources, and the different mechanisms which various fish employ in the incorporation/elimination of the congeners. Therefore, one should not use 2,3,7,8-TCDD data to estimate the TE value for a given fish.Also, based upon the coefficients of skewness and kurtosis, the distribution of the differences between the 2,3,7,8-TCDD concentrations and the calculated TE values was not a normal distribution. One must, therefore, be careful in applying statistical tests to the data. The more routinely used statistical parameters such as t-tests and F-tests cannot be correctly used for nonnormally distributed populations. 419
