Among environment scientists, the current and almost universally accepted definition of dissolved constituents is an operational one; only those materials which pass through a 0.45-??m membrane filter are considered to be dissolved. Detailed laboratory and field studies on Fe and Al indicate that a number of factors associated with filtration, other than just pore size, can substantially alter 'dissolved' trace element concentrations; these include: filter type, filter diameter, filtration method, volume of sample processed, suspended sediment concentration, suspended sediment grain-size distribution, concentration of colloids and colloidally associated trace elements and concentration of organic matter. As such, reported filtered-water concentrations employing the same pore size filter may not be equal. Filtration artifacts may lead to the production of chemical data that indicate seasonal or annual 'dissolved' chemical trends which do not reflect actual environmental conditions. Further, the development of worldwide averages for various dissolved chemical constituents, the quantification of geochemical cycles, and the determination of short- or long-term environmental chemical trends may be subject to substantial errors, due to filtration artifacts, when data from the same or multiple sources are combined. Finally, filtration effects could have a substantial impact on various regulatory requirements.