Methane is the second most important greenhouse gas for which the oceans are a small but significant source. Coastal waters are usually very rich in methane, but the dynamics of production, consumption and distribution is only vaguely understood. In this project, the seasonal distribution of methane in the temperate coastal water column was studied in 1996 and 1997 at two contrasting sites off the Isle of Anglesey: the sheltered Menai Strait and the exposed area at Point Lynas. The study was based on methane concentration measurements conducted by headspace equilibration followed by gas chromatography. Although concentrations were 2-to 10-fold highe r at the more sheltered location, the waters at both sites were supersaturated with methane relative to the atmosphere throughout the year (saturation ratios from 1.0 to 26.4) -which indicates a continuous source of supply. Following a fortnightly sampling programme both sites were found to disp lay an increase in methane concentrations with temperature in the summer. However, high short-term variability in methane concentrations precludes confirmation of a 'seasonal' effect. In a n attempt to explain the distribution of water column methane in space and time potential sources and sinks were investigated for the Menai Strait. Advection of water from Liverpool Bay could account for as much as 20% of the concentration measured in the central Menai Strait, while other external sources such as the River Ogwen and treated and raw sewage discharges were only minor contributors. It was concluded that most of the Menai Strait methane was autochthonous. Methane production in the water column was only detected on one occasion. During a bloom of the heterotrophic dinoflagellate Noctiluca scintillans low methane produc tion rates were observed during incubation experiments. This was the first recorded association between Noctiluca and methane production. It is speculated that methane is produced by bacteria that have previously been described to occur within Noctiluca cells (Lucas, 1982). Even during dense blooms of Noctiluca, however, the internal water column source for methane was calculated to be quantitatively insignificant for coastal zones. By a process of elimination, muddy sediments were identified as the main source for methane in the water column, but the calculated diffusive flux rates were too low to account for all the methane that must be produced in the Menai Strait. The diffusive flux of methane from the sediment into the water showed no seasonal pattern. The atmosphere is the main sink for methane from the water column. Wind was negatively correlated with methane concentrations and could account for much of the short-term variability in water column methane concentrations. The atmospheric sink strength was estimated with three wind velocity-dependent gas exchange models. The discrepancies in estimates from different flux models were significant. Flux estimates varied by a factor of up to 7.5, with lower discre pancies at higher wind speeds. Although wind speeds were observed to be generally lower during the summe r, more methane was calculated to evade the water in the summer than the winter. This suggests that the high concentrations in the summer water column must be caused primarily by increased production during the summer, and that the seasonal pattern observed in the water column is likely to be real. The fluxes to the atmosphere from Point Lynas and the Menai Strait were similar, suggesting that, despite large differe nces in concentrations, the source stre ngths at the two sites were also similar. The flux estimates demonstrated the inadequacy of existing gas exchange models as they produced different answers to the question of which of the two studied sites exported more methane to the atmosphere. Although the broad features of the dynamics of methane in the study area have been characterised it was not possible to close the mass balance budget.