The existence of two IgA subclasses in humans has been reliably shown by biochemical, immunochemical and genetic means. IgA is unique among immunoglobulins in the regular occurrence of both monomeric and polymeric forms. In order to be able to study the relationship between monomeric and polymeric IgA1 and IgA2 concentrations in the circulation and mucosal compartment i.e. secretions, it is essential that the methods used are not biased by the molecular size of the IgA under investigation. We validated IgA and IgA subclass measurements in serum and saliva by sandwich enzyme-linked immunosorbent assay (ELISA). Coating reagents were specific mAbs against IgA (clone 4E8), IgA1 (clone 69-11.4) or IgA2 (clone 16-512-H5 and clone IF8.58). Pooled normal human serum and purified dimeric IgA1 (d-IgA1) or IgA2 (d-IgA2) myeloma proteins were used to standardize the assays. Polymeric and monomeric forms of IgA in sera from volunteers and patients with myelomatosis were assayed in fractions separated by high performance liquid chromatography (HPLC). Dithioerythritol (DTE) was used to determine the influence of the quarternary structure of IgA on its detection by mAbs. We found that mAbs 4E8, 69-11.4 and 16-512-H5 reliably measured d-IgA, d-IgA1 and d-IgA2 respectively, independent of the standard employed. Clone IF8.58 underestimated the concentration of d-IgA2 (correction factor +/- 2) with increased sensitivity in the presence of DTE. This difference is probably explained by the composition of the immunogen against which the mAb was raised. We conclude that no reliable conclusions can be made concerning the subclass ratio in biological fluids unless the monoclonal antibodies used have been appropriately validated.