Abstract
We have compared, using DNA analysis, the classical `pull and shake' method for sampling rhizosphere soil with a model rhizosphere system in which the soil was sampled at increasing distances (0.5 mm) from the root mat of Brassica napus. These studies showed that more DNA was extracted from the soil slices closest to the roots. Whilst the gradient in DNA recovered cannot be attributed solely to differences in microbial biomass, it clearly illustrates the lack of spatial resolution characteristic of the `pull and shake' method of sampling from the rhizosphere. The DNA-rich fraction (rhizosphere soil) was compared with bulk soil DNA by PCR using actinomycete 16S rDNA primers and 16S rDNA sequencing. Phylogenetic analyses of 16S rDNA sequences showed that these primers were not specific for actinomycetes but also amplified Gram positive low G+C bacterial sequences. Comparative 16S rDNA analysis of these sequences showed an enrichment for Gram positive low G+C bacteria in the rhizosphere of Brassica napus.