ABSTRACT

In ‘‘traditional’’ microbiology, soil bacteria are quantified by methods that detect viable cells,

by plate counting or most probable number (MPN) enumeration, for example. The major

weakness of this approach is that only those organisms that are viable and able to grow in the

chosen media at the specified incubation conditions (e.g., temperature) will be detected.

Microbiologists have long suspected that the bacteria that are amenable to culturing using

conventional methods represent only a tiny fraction of those in soil. Most likely, because these

have fastidious requirements that have foiled the development of suitable cultivation tech-

niques, or because they are in obligate association with other organisms such as protozoa. The

reannealing behavior of DNA isolated from soil suggests that a single gram of soil may contain

up to 10,000 bacterial types (Torsvik et al. 1996). A gram of agricultural soil typically contains

a billion or more bacteria. There are estimated to be about 51030 prokaryotic individuals on Earth, of which 491027 are in the top meter of cultivated land on the Earth’s surface (Whitman et al. 1998). The exploration of this hitherto unseen microbial world is now feasible

using methods that can elucidate the abundance, identity, and activity of bacteria without

relying on culturing. Methods that exploit the sequence of bacterial nucleic acids extracted

directly from soil are particularly powerful in this regard.