neering bacteria to produce more reliable results, as well as expanding the range of bioremediation to include areas contaminated by metals, pesticides, radioactive elements, and mixed wastes.

One of the biggest roadblocks to development is that so little is known about bacterial communities in nature. Although microbes are the most abundant and widespread organisms on earth, their ecology is largely a mystery. The immediate need is to discover how microbial communities function in the wild, and how they respond naturally to stresses, such as exposure to materials that are toxic to most organisms.

Using microbes to carry out tasks in the great outdoors poses a practical difficulty: How on earth do you keep track of what they're doing? It's especially challenging when the microbes are working below the surface of the soil—for example, when they are breaking down underground contaminants. One method of monitoring that has been widely tested involves linking the genes that cause bacteria to degrade contaminants with genes for producing bioluminescence. Bioluminescence is biologically produced light, made, for example, by fireflies, glowworms, some fungi, and many marine organisms. The result of this genetic linking is that the bacteria light up whenever they are working at decontamination.

In experiments already carried out, genes for light production (lux genes) have been coupled to a microbe's genes for naphthalene degradation. The microbe's activity is measured on-site by changes in light level recorded