Automated conductance measurements in polypectate medium were used for the detection of pathogenic soft rot Erwinia spp. in potato peel extracts. The detection threshold for Erwinia carotovora subsp. atroseptica (Eca) in inoculated peel extracts was ca. 104 colony forming units (cfu) ml-1 when samples were considered positive on the basis of a response within 48 h at 20 °C. Detection of E. chrysanthemi (Ech) was less sensitive, only 105 cfu ml-1 peel extract were detected within 36 h at 25 °C. The linear correlation between detection times in conductimetry and inoculum levels of Eca and Ech in peel extracts was used for a quantitative estimation of Eca and Ech in naturally contaminated peel extracts. Samples giving a positive conductimetric response had to be confirmed with an enzyme-linked immunosorbent assay (ELISA) or polymerase chain reaction (PCR) for the presence of Eca and Ech, because E. carotovora subsp. carotovora (Ecc) also generated a conductance response. Conductimetry was sensitive and efficient for detection of contamination levels of Eca higher than 104 cfu ml-1 peel extract. For Ech, conductimetric detection was less sensitive and inefficient due to low contamination levels of Ech and the presence of high numbers of Ecc in many samples after enrichment, which interfered with the test. Immunofluorescence cell staining (IF) combined with enrichment and immunofluorescence colony staining (IFC) were suited to detect and quantify low numbers of Eca and Ech at less than 104 cfu ml-1 in peel extracts. However, since false positive and negative reactions in serology were observed, the use of PCR after enrichment, or in combination with IFC to confirm positive results, was required for accurate detection.
