Selenium has aroused much interest in the last decade. It is an essential trace element for man and animals and has been found to be the cofactor in two enzyme systems. The first, glutathionine peroxidase, acts as an anti-oxidant by destroying peroxides which attack cellular membranes. The second, iodothyronine 5'-deiodinase, converts thyroxine to triiodothyronine with the release of iodine. Selenium is obtained in the diet from plant and animal products and the amount present is dependent largely on the Se content and the chemical species present in local soils. Soils low in Se can lead to deficiency problems to the grazing ruminants, and have necessitated the use of Se fertilisers or direct Se injection into the livestock. The disposal of sewage sludge onto agricultural land will increase as new EC legislation comes into force in 1998. This may alleviate Se deficiency in soils or result in toxicity problems. The aims of this project were firstly to develop and validate a method to accurately measure Se in plants, soils and sewage sludges, examine which soil fractions Se was associated with and finally to study the uptake of Se by plants grown on sewage sludge amended soils. The Se concentration in samples was determined using isotope dilution - mass spectrometry by gas chromatography - mass spectrometry (GC-MS). This was the first application of the determination of Se in environmental samples and the use of a benchtop GC-MS for those analyses. Samples were spiked with 76-Se isotope solution. Plants were digested using nitric acid and hydrogen peroxide initially on a heating block and latterly by microwave oven digestion techniques. Soils and sewage sludges were digested using nitric acid and hydrofluoric acid. Selenium in the digests was reduced to Se(IV) and derivatised to 5'-nitropiazselenol. Validation of the methodology was achieved by the use of certified reference materials which gave results within the certified range with a low standard deviation. The Se content of four different freely drained acid Scottish soils under grasslands was in the range 0.5-0.8 mg-1 air dried soil and three sewage sludge samples contained between 1.1 and 3.5 mg g-1 dry matter. The ability of the digestion techniques to release organically bound Se as selenomethionine and selenocysteine was found to be around 100% efficient and the use of a microwave oven increased the number of samples which could be processed. A sequential extraction procedure was used to determine the soil fractions with which Se was associated. The four soils and the sewage sludge examined were all found to have less than 5% of the total Se available for plant uptake. The uptake of Se by plants grown in soils to which Se had been added as sodium selenate or sodium selenate was studied using both stable and radio isotopes of Se. Results obtained showed that the uptake of Se by plants grown in soils treated with selenite could be related to the soil characteristics, with ferric oxide levels, clay levels and organic matter content being the important factors. The uptake of Se by plants in the soils treated with selenate were less obviously related to soil properties but pH and P levels both appear to be important factors. Good reproducibility was obtained in all cases. Plants grown on sewage sludge amended soils were not found to contain measurable amounts of Se.