Within the quality control scheme of the Deutsche Gesellschaft für Klinische Chemie, seven quality control surveys of thyrotropin (TSH) determinations in blood dried on filter paper have been carried out since 1980. Ninety-six screening laboratories from 12 European countries took part in these surveys. In a single survey each participant usually analysed four different samples; each of these consisted of three spots of dried blood spiked with defined amounts of thyrotropin. For the evaluations of the surveys the participants were asked to give information about their analytical results, and from these, their diagnostic classifications. The medians of the analytical results correlated well with the given thyrotropin concentrations, but the individual estimations from different laboratories varied greatly. Major discrepancies of classification were also apparent, both in the given thyrotropin concentrations and in the individual estimations. Two special collaborative studies with nine selected laboratories showed on the one hand that analysis of the largest possible part of the dried blood sample can help to optimize the precision of the results; on the other hand, considerable drawbacks related to the reagents and the methods were sometimes observed.