An integrated approach to gas analysis in soil cores was conducted to provide a novel method for observing the gas dynamics associated with upland soil ecosystems. Depth profiles of the O(2), Ar, CO(2), CH(4), N(2) and NO(x) concentrations in intact soil monoliths were obtained simultaneously using membrane inlet mass spectrometry (MIMS). This technique enables the direct measurement of multiple gas species throughout the soil core with minimal disturbance. Depth profiles provided data on the vertical heterogeneity of gas concentrations, while horizontal heterogeneity was monitored by comparison between profiles. Detailed descriptions of the modifications to current MIMS methods for in situ environmental monitoring of terrestrial soils are provided. These included a thorough examination of calibration of the MIMS probe in gas phase, stirred and unstirred H(2)O, or between glass beads immersed in H(2)O. Calibration was also carried out in sterile (autoclaved) soil. The mean concentrations of CO(2) and CH(4) in the soil monoliths increased from 27 microM and undetectable levels respectively at the surface, to maximum values of 3.6 mM and 4.3 microM at 12-cm depth. These changes corresponded with decreases in mean O(2), Ar and N(2) concentration from 300, 20 and 720 microM respectively to 0-6, 10 and 574 microM at 12-cm depth. These data indicated the presence of a gradient within the core from an aerobic environment to an O(2)-depleted, but not in all cases a completely anaerobic, one. This transition corresponded, to some extent, with that between the upper and lower soil horizons. The increased methane and CO(2) concentrations observed at depth are indicative of anaerobic environments. General trends associated with the gradually changing vertical heterogeneity of these gas profiles and the transition towards anaerobiosis did not provide evidence for the existence of localised microsites. Some evidence for microsite-specific microbial communities was however, provided by observation of broad zones of accumulation of NO(x) species, but only at concentrations close to the limit of detection of the method. The ratio of each gas, to argon was calculated at each depth. This was done to correct for physical parameters, which influence inert and biologically active gases, equally. The amount of di-nitrogen as a ratio to Ar was seen to increase with depth. This could be evidence for denitrification in the lower horizon. An example of the dynamic 'online' data collection capabilities is provided for diurnal oscillations in subsurface (5 cm) soil gas concentrations.