Sir, Plasmid-mediated quinolone resistance was first reported in 1998, in a Klebsiella pneumoniae isolated in 1994 in Alabama, USA. Since then, a number of surveys have reported an increasing prevalence of qnr genes, starting in the mid-1990s, and the appearance of aac(60)-Ib-cr from 2000 onwards. The oldest qnr genes found so far are from isolates obtained in 1988 (a qnrB8-like gene in a Citrobacter freundii from the USA and a qnrB9-like gene in a K. pneumoniae from Argentina); a qnr-positive Enterobacter isolate from 1994 has also been found in Israel. The earliest reported qnr genes from the Nordic countries are from 2002, while a recent paper from Sweden found qnr-positive ESBL strains from 2006 onwards in a collection spanning the years 2001–08. Most surveys have studied clinical isolates, although it has been observed that the qnr genes presumably have been circulating for some time, due to their appearance in various genera on all continents. The surprisingly high prevalence of qnr genes found in faecal enterobacteria (54% contained qnrB) in healthy children in South America shows the importance of commensal bacteria as a resistance gene pool. We decided to screen our older Enterobacteriaceae collections from the 1990s for the presence of qnr genes. The collections consisted of 365 faecal isolates, which were collected during research projects studying resistance levels in the human faecal microbiota. These consisted of 227 isolates from 1993– 94, from healthy persons living in Turku, Finland, who had not received antibiotics for 3 months before sampling, and 138 isolates from long-term patients collected in 1994 (mean hospitalization time 22 months). We also included 561 isolates from vegetables (fresh and frozen) isolated in 1996–97. Isolates with ciprofloxacin MICs of 0.125 mg/L were selected, and 43 such isolates were found: one from the healthy group; two from the patient group; and the rest from the vegetable collection. Of these, 23 were recovered from storage. To prepare DNA, bacteria were cultured overnight and a colony was suspended in sterile water and lysed by heating at 958C for 10 min. Screening for qnrA, qnrB and qnrS genes was carried out by multiplex PCR using a previously described method and specific primers. Any qnr-positive results were confirmed by direct sequencing of both strands of amplicons using specific PCR primers. Sequencing reactions were performed using an ABI BigDye Terminator Cycle Sequencing Kit version 3.1 (Applied Biosystems, Espoo, Finland) and sequenced with an Applied Biosystems 3730 DNA Analyser. The DNA sequences were analysed and translated into amino acid sequences with Vector NTI software (Informax, Inc., No. Bethesda, MD, USA). The closest matches to these amino acid sequences were then obtained using the BLAST search engine (http://www.ebi.ac.uk/Tools/blast/). One isolate was positive for qnrB. Sequencing showed it to be qnrB12, according to the Jacoby et al. nomenclature. The isolate was a Citrobacter youngae, as determined by API20E (bioMerieux, France), isolated on 29 June 1994 from a healthy person .30 years old. The qnrB12-positive strain was susceptible to all antibiotics (Table 1), with the possible exception of co-amoxiclav (the activities of the intrinsic b-lactamases of the Citrobacter genus vary and there is not much published on C. youngae). qnrB12 has been reported once before, in three German Citrobacter werkmanii poultry isolates from the years 1999, 2002 and 2005. The gene was chromosomally located in these strains. This is the first report of qnr genes from Finland and the oldest reported so far in Europe, and, indeed, is among the oldest also internationally. The fact that it is not a clinical strain corroborates the observation that these genes have been circulating widely for a long time. Since the strain is susceptible by both CLSI and EUCAST (‘European Committee on Antimicrobial Susceptibility Testing’) breakpoint criteria, the presence of the gene would pass unobserved until further mechanisms were acquired.