formation by individual cells on agar growth medium provided estimates of microbial population sizes. Twenty years ago, microbiologists had described only 5,000 kinds of bacteria and archaea, and biodiversity studies largely ignored the contribution of single-cell organisms to Earth’s global biomass. These modest assessments of microbial diversity and population size were not consistent with a ~3.5 billion-year evolutionary history, during which single-cell organisms have developed an enormous metabolic repertoire to cope with Earth’s dynamic environment. Such underestimates of microbial diversity reflect how difficult it is to identify morphological and biochemical characteristics (phenotypic traits) that are uniquely shared between closely related organisms, as opposed to common ancestral features that persist over the largest of evolutionary timescales.

Discoveries over the last 20 years of microbial life in habitats previously thought to be devoid of life (e.g., hot springs, deep-sea vents, solid ice, subglacial environments), and the ability to detect microorganisms without requiring their cultivation in the laboratory, have greatly expanded our knowledge of habitat range and numbers of microbes in Earth’s biosphere. For example, fluorescent DNA staining of bacterial cells and epifluorescence microscopy revealed that microbial numbers in aquatic environments were 100 to 1,000 times greater than estimates based on cultivation techniques.4 Contemporary culture-independent surveys demonstrate that microbes may be the dominant biomass on Earth (Whitman et al., 1998), and microbes account for most of the genetic and metabolic diversity (Pace, 1997). The number of different kinds of microbes remains unknown, but the use of modern molecular technology indicates that traditional estimates of 5,000 kinds of microbes are low by several orders of magnitude. In addition to high diversity, the number of archaean and bacterial cells on Earth (see Table 5.1) exceeds 3.6 × 1030, while viral and bacteriophage titers are several orders of magnitude higher. These microbial abundances yield a total carbon content that is 50 percent or more of the estimated carbon in all terrestrial plants (Whitman et al., 1998).


The direct interrogation of microbial genomes offers a powerful, culture-independent method for exploring microbial diversity. Every species has a unique collection of gene sequences that make up its genome, and each species can have a very different complement of genes. The collection of gene sequences in a genome specifies the constellation of proteins that orchestrate cellular biochemistry. The aggregate biochemical activity defines the character or phenotype of a particular kind of organism. The DNA sequence of a gene can serve as a proxy for species identification if that gene is present and conserved by evolution in both the unknown and a reference set of well-characterized organisms. The comparison of gene sequences that share a common evolutionary history allows the reconstruction of evolutionary history or phylogeny. When an organism of unknown taxonomic affinity is included in such analyses, computer algorithms can infer its phylogenetic placement in the context of a large molecular database of well-characterized lineages.


comparison of morphological and physiological traits to differentiate between microbial species. Molecular techniques including DNA:DNA hybridization, determination of GC content of the genome (percentage of the nucleotides containing guanine and cytosine), and phylogenetic inferences based on comparisons of small-subunit ribosomal RNA gene sequences provided more powerful tools for differentiating among microbes. Yet an attempt to define species boundaries according to a defined number of nucleotide differences between homologous genes is arbitrary and in some cases fails to resolve different populations in situ (e.g., Ward et al., 1990).


A serial dilution assay is a common method for enumerating numbers of bacteria by counting the number of colonies formed by aliquots of sample dilutions spread onto an agar surface. A small known amount of sample is mixed with a sterile diluent solution (water or liquid media), and a 0.1- to 1.0-ml aliquot of the mixed diluent (containing the initial sample) is spread on an agar surface. In a similar way, successive dilutions are prepared from each diluent and 0.1- to 1.0-ml aliquots are spread onto an agar plate. After counting the number of bacteria colonies that grow on each plate, it is possible to backcalculate, using the “dilution factor” (the number of times that bacteria sample was diluted with the diluent solution), the number of bacteria in the original sample.


Epifluorescent microscopy relies on the excitation of susceptible molecules in a sample with short-wavelength, high-energy light and observation of the emitted lower-energy light (fluorescence). Susceptible molecules include autofluorescent compounds such as chlorophyll and F420 (an enzyme cofactor found in methanogens) or fluorophores that bind to specific cellular compounds. Epifluorescent microscopy enables researchers to observe and detect cellular components that were not visible with conventional light microscopy, and it is sensitive enough to detect a single molecule.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement