The prevalence of selected streptomycin (Sm)-resistance genes, i.e. aph (3''), aph (6)-1d, aph (6)-1c, ant (3'') and ant (6), was assessed in a range of pristine as well as polluted European habitats. These habitats included bulk and rhizosphere soils, manure from farm animals, activated sludge from wastewater treatment plants and seawater. The methods employed included assessments of the prevalence of the genes in habitat-extracted DNA by PCR, followed by hybridisation with specific probes, Sm-resistant culturable bacteria and exogenous isolation of plasmids carrying Sm-resistance determinants. The direct DNA-based analysis showed that aph (6)-1d genes were most prevalent in the habitats examined. The presence of the other four Sm-modifying genes was demonstrated in 58% of the tested habitats. A small fraction of the bacterial isolates (8%) did not possess any of the selected Sm-modifying genes. These isolates were primarily obtained from activated sludge and manure. The presence of Sm-modifying genes in the isolates often coincided with the presence of IncP plasmids. Exogenous isolation demonstrated the presence of plasmids of 40-200 kb in size harbouring Sm-resistance genes from all the environments tested. Most plasmids were shown to carry the ant (3'') gene, often in combination with other Sm-resistance genes, such as aph (3'') and aph (6)-1d. The most commonly found Sm-modifying gene on mobile genetic elements was ant (3''). Multiple Sm-resistance genes on the same genetic elements appeared to be the rule rather than the exception. It is concluded that Sm-resistance genes are widespread in the environmental habitats studied and often occur on mobile genetic elements and ant (3'') was most often encountered.

Transfer of plasmid RP4 between introduced strains of Pseudomonas fluorescens was studied in 2 soils, Ede loamy sand and Guelph loam, in non‐rhizosphere and rhizosphere soil using soil chambers and microcosm systems. Short‐term organism survival was generally at high levels (> 106/g dry soil), in both soils, whereas long‐term survival was poorer, particularly in the loamy sand. Amendment of this soil with bentonite clay improved bacterial survival. Plasmid transfer between donor and recipient strains freshly introduced into separate portions of Ede loamy sand, which were subsequently mixed, was only detected in the vicinity of growing wheat roots, suggesting roots stimulate bacterial migration and/or growth. However, no transfer was detected between resident donor and recipient cell populations (introduced 48 days previously), due to poor organism survival. Plasmid transfer was detected in the rhizosphere between established, resident donor cell populations, and newly‐introduced recipients, and vice‐versa, in both soils. These data suggested that plant roots enhance the frequency of bacterial matings not only between organisms present in the same niches, but also between organisms from different niches, or in different conditions of stress, probably by stimulating bacterial migration and/or growth, or by providing additional surfaces for cell‐to‐cell contact.