Prior to 1995, only limited and fragmented data were available to assess the likelihood that a suspect might have become innocently contaminated with traces of high explosives through contact with the general public environment. This second survey to determine the background levels of explosive traces in public places focused on public means of transportation, as did the first survey, but also included samples collected from hotel rooms, private houses, private vehicles, and clothing. In order to ensure consistent and easy sampling of a variety of sites, standardized sampling kits were built for each type of site. This paper describes the content of the sampling kits, quality assurance of the kits, the sampling of the chosen areas, control samples, the processing of the samples, the cleanup process, the analysis of the samples, the interpretation of chromatograms, and the analysis of the samples for the presence of HMX (high explosive species cyclotetramethylene tetranitramine). The survey found that traces of the high explosive nitroglycerine (NG), trinitrotoluene (TNT), pentaerythritol tetranitrate (PETN), and cyclotrimethylene trinitramine (RDX) were rare within the general public environment. Only one low-level trace of RDX was detected. NG, possibly associated with the use of firearms, was detected at low levels in two samples, and 2,4-DNT was detected in a separate sample. No PETN was detected in any of the samples. The analytical procedures used would also have detected ethylene glycol dinitrate (EGDN) if present at levels greater than 2 ng, nitrobenzene (NB) if present at levels greater than 50 ng, mononitrotoluenes if present at levels greater than 50 ng, and the other common isomers of dinitrotoluene if these had been present at levels in excess of 10 ng. None of these were detected. Approximately 7 percent of the sample were analyzed for the presence of HMX; no HMX was detected. Thus, the survey findings indicate that it is unlikely that persons who visit public areas in the four major cities could become significantly contaminated with explosives; however, further work at more sites and different types of sites would increase the value of the data. 12 tables, 3 figures, and 2 references