We used the PCR to study the presence of two plant pathogens in archived wheat samples from a long-term experiment started in 1843. The data were used to construct a unique 160-yr time-series of the abundance of Phaeosphaeria nodorum and Mycosphaerella graminicola, two important pathogens of wheat. During the period since 1970, the relative abundance of DNA of these two pathogens in the samples has reflected the relative importance of the two wheat diseases they cause in U.K. disease surveys. Unexpectedly, changes in the ratio of the pathogens over the 160-yr period were very strongly correlated with changes in atmospheric pollution, as measured by SO 2 emissions. This finding suggests that long-term, economically important, changes in pathogen populations can be influenced by anthropogenically induced environmental changes.Mycosphaerella graminicola ͉ quantitative PCR ͉ Phaeosphaeria nodorum ͉ septoria diseases ͉ sulfur

Summary• Relationships between weather, agronomic factors and wheat disease abundance were examined to determine possible causes of variability on century time scales.• In archived samples of wheat grain and leaves obtained from the Rothamsted Broadbalk experiment archive (1844 -2003), amounts of wheat, Phaeosphaeria nodorum and Mycosphaerella graminicola DNA were determined by quantitative polymerase chain reaction (PCR). Relationships between amounts of pathogens and environmental and agronomic factors were examined by multiple regression.• Wheat DNA decayed at approx. 1% yr -1 in stored grain. No M. graminicola DNA was detected in grain samples. Fluctuations in amounts of P. nodorum in grain were related to changes in spring rainfall, summer temperature and national SO 2 emission. Differences in amounts of P. nodorum between grain and leaf were related to summer temperature and spring rainfall. In leaves, annual variation in spring rainfall affected both pathogens similarly, but SO 2 had opposite effects. Previous summer temperature had a highly significant effect on M. graminicola. Cultivar effects were significant only at P = 0.1.• Long-term variation in P. nodorum and M. graminicola DNA in leaf and grain over the period 1844-2003 was dominated by factors related to national SO 2 emissions. Annual variability was dominated by weather factors occurring over a period longer than the growing season.