A conjugation assay was used to determine the effects of phenotypic resistance to one to up to 5 antibiotics, sampling site of origin, presence or absence of class 1 and/or class 2 integrase (intI) genes (intI1 and intI2), and the number of sulfamethoxazole resistance (sul) and trimethoprim resistance (dfr) genes on the transfer frequencies of plasmids from environmental, antibioticresistant Escherichia coli. Of 51 sulfamethoxazole and trimethoprim-resistant E. coli isolates conferring at least one mob gene (mob P51 , mob F11 , mob F12 , mob Q11 , mob Q12 , or mob Qu ), 38 produced transconjugants with an overall mean frequency of 1.60 £ 10 ¡3 transconjugants/ donors (T/D) or 5.89 £ 10 ¡3 transconjugants/recipients (T/R). The presence or absence of intI1 and intI2 and the presence or absence of different targeted dfr genes (dfrA1, dfrA8, dfrA12, dfrA14, dfrA17, and/or dfrB3) were not statistically related to plasmid transfer frequencies as determined by ANOVA (P 0.05). However, E. coli isolates recovered 2 km downstream of wastewater treatment plant effluent input, and those possessing resistance to 3 antibiotics had significantly greater plasmid transfer frequency than their counterparts when calculated as T/D (ANOVA followed by Fisher's least significant difference means comparison, P < 0.05). Greater plasmid transfer frequency calculated as T/D was also measured for E. coli possessing 3 compared to a single sul gene. The invitro frequency suggests that horizontal gene transfer of conjugative mediated-antibiotic (sul) resistance genes may be significant among resistant, stream bacteria.