The normal microbial flora of the vagina plays an important role in preventing genital and urinary tract infections in women. Thus an accurate understanding of the composition and ecology of the ecosystem is important to understanding the aetiology of these diseases. Common wisdom is that lactobacilli dominate the normal vaginal microflora of post-pubertal women. However, this conclusion is based on methods that require cultivation of microbial populations; an approach that is known to yield a biased and incomplete assessment of microbial community structure. In this study cultivation-independent methods were used to analyse samples collected from the mid-vagina of five normal healthy Caucasian women between the ages of 28 and 44. Total microbial community DNA was isolated following resuspension of microbial cells from vaginal swabs. To identify the constituent numerically dominant populations in each community 16S rRNA gene libraries were prepared following PCR amplification using the 8f and 926r primers. From each library, the DNA sequences of approximately 200 16S rRNA clones were determined and subjected to phylogenetic analyses. The diversity and kinds of organisms that comprise the vaginal microbial community varied among women. Species of Lactobacillus appeared to dominate the communities in four of the five women. However, the community of one woman was dominated by Atopobium sp., whereas a second woman had appreciable numbers of Megasphaera sp., Atopobium sp. and Leptotrichia sp., none of which have previously been shown to be common members of the vaginal ecosystem. Of the women whose communities were dominated by lactobacilli, there were two distinct clusters, each of which consisted of a single species. One class consisted of two women with genetically divergent clones that were related to Lactobacillus crispatus, whereas the second group of two women had clones of Lactobacillus iners that were highly related to a single phylotype. These surprising results suggest that culture-independent methods can provide new insights into the diversity of bacterial species found in the human vagina, and this information could prove to be pivotal in understanding risk factors for various infectious diseases.