We found that levels of antiphospholipid antibodies (aPLA), measured with an ELISA increase if serum or plasma samples are heated. The phenomenon is dependent on duration and degree of heating, optimum levels being reached at 3 h at 56 degrees C. Negative samples become positive after heating. The heating effect is more pronounced for IgG-aPLA than for IgM-aPLA and is not observed for adsDNA, atetanus or lymphocytotoxic antibodies. The presence of serum/plasma components in addition to IgG is essential for the phenomenon to occur. Ultracentrifugation and mixing experiments with isolated IgG did not enable us to explain our observations. Nevertheless, knowledge of this phenomenon is of practical importance.