During the last five years, work has been underway at the Lawrence Livermore National Laboratory (LLNL) to develop a method for imaging gas clouds that are normally invisible to the human eye. The effort was initiated to provide an effective means of locating leaks of hazardous vapors. Although conventional point or line-of-sight detectors are well suited to the measurement of gas concentrations, their utility in identifying the origin and direction of travel of gas plumes is limited. To obtain spatial information from sensors that provide only zero- or one-dimensional readings, either sequential readings at many different locations from a single device, or multiplexed simultaneous measurements from a sensor array must be taken. The former approach is time consuming and, therefore, impractical in emergency situations where rapid action is required. The latter is useful only in cases where the probability of a hazardous release is high enough to warrant the prior installation of a sensor network. Either method demands high measuremental precision and sufficient discrimination against both interfering gases and interfering sources of the target gas. Backscatter Absorption Gas Imaging (BAGI) is a new technique that makes gas clouds and their surroundings "visible" in a real-time video image. It is superior to conventional sensors in characterizing the spatial properties of gas clouds because it provides data that are inherently two-dimensional. Less measuremental precision is required by the BAGI technique because it conveys information as contrasts between different areas in an image rather than as absolute concentration values. Furthermore, the pictorial display of this information allows it to be rapidly assimilated by emergency-response teams. The size and orientation of the plume are evident through comparison with familiar objects that also appear in the image. Subtler evaluations can be made as well, such as the distinction between innocous and hazardous sources of the target gas. For example, in using a conventional sensor to search for the source of a gas that is also present at low levels in automobile exhaust, one might be led astray near a highway. Gas imaging allows the searcher to recognize that the cars are producing the gas, and that they are not the objective of the search.