Soils are the most important source of N₂O emissions to the atmosphere, with denitrification and nitrification being the major processes responsible for production of the gas. Although much is known about factors controlling these processes and N₂O fluxes from them it is still difficult to obtain accurate N₂O emission estimates due to the highly heterogeneous nature of soils. Better estimates can only be achieved by a combination of direct measurements in key ecosystems and by quantifying the relationships between fluxes and the controlling parameters, as an aid to modelling and upscaling. The aim of this project was to quantify the effects of various soils and environmental parameters on N₂O emissions to the atmosphere, making measurements in a semi-controlled environment, where it was more possible to control these parameters than in the field. The system consisted of 12 soil monoliths (1 m diameter and ca. 60 cm deep) from three contrasting soils (a sandy loam, a clay loam and a peaty gley). The headspaces of the monolith casings were converted to flux chambers by fitting them with aluminium lids and each chamber was connected to an ECD gas chromatograph. Gas sampling and analysis, and recording of information from temperature probes and transducer tensiometers, were completely automated. Soil water content, temperature (including diurnal temperature variation), organic matter input and respiration all had major effects on N₂O emissions. Using boundary line analysis (summarising data from several experiments), quadratic relationships between water-filled pore space (WFPS) and log-transformed N₂O fluxes from the sand loam and clay loam soils were established; the optima for emission were 90 and 92% WFPS, respectively. The relationships between temperature and log-transformed N₂O data were linear, and Q₁₀-values up to 7.5 for the sandy loam soil and 9.4 for the clay loam soil were observed. The high optimum WFPS for emissions and the high Q₁₀-values indicate that denitrification was the major process involved. Diurnal maxima in N₂O flux were observed, which sometimes coincided with the temperature maxima in the uppermost 5 cm, but on other occasions the flux maxima were delayed by several hours; this was attributed to N₂O production taking place at greater depths. Significant relationships were observed between N₂O emissions, and CO₂ emissions from respiration, following incorporation of a grass-clover mixture into the sandy loam and clay loam soils. The overall effect of respiration on log-transformed N₂O emissions from the sandy loam and clay loam soils could be described with a rectangular hyperbola, where the rate of the N₂O emission increase at first rose steeply with the respiration rate, but then slowed down drastically when the respiration rate was greater than 20 mg CO₂-C m^-2 h^-1. No boundary line could be defined for water-filled pore space, temperature and respiration from the peaty gley soil. However, when data from single experiments were analysed, relationships could be established. Strong interactions between all the factors controlling N₂O emissions were observed.