Most plasmids in natural populations tend to be on the larger site, not very many are in the 2-10kb range (although there certainly are some). The larger plasmids, especially the ones that can conjugate, may be 50-200kb and hence will probably be indistinguishable from the genomic DNA. Small plasmids are likely to also be present but they may not be readily visible as they are likely to be a very small population of the total if this is truly a mixed environmental sample.
So what is the basis that makes you think you don't have plasmids in your DNA prep?
While isolating plasmid DNA from direct sample using modified Brinboim and Dolly's protocol, I got the genomic DNA in high concentrations rather than plasmid DNA, in most of the cases. Sometimes plasmids were visible as very thin feeble bands in the gDNA smear. if your purpose is to go for plasmid based PCR amplification, then this would help, but if you want to go for sequencing and all, i would suggest culture based plasmid isolation