Samples of inedible bulk mutton tallow were collected monthly throughout two killing seasons from one meat killing plant. These samples plus one sample from another plant, were analysed for fatty acid and triglyceride composition. In these samples, four fatty acids (myristic (14:0), palmitic (16:0), stearic (18:0) and oleic (18:1)) comprised 88.6% of the total fatty acids, and there was an average of 16% trisaturated triglycerides, 38% disaturated triglycerides and 46% of triglycerides with a greater degree of unsaturation. Overall, there was a significant decrease in the proportion of 14:0, and a significant increase in the proportion of 18:0, from November to June; and there was a significant difference in the mean proportion of 16:0, and also 18:1, between the two seasons. There was a significant difference in the proportion of cis monounsaturated triglycerides, and the more highly unsaturated triglycerides, between some of the different tallows analysed. There was a significant decrease in the proportion of 2-oleo disaturated triglycerides from November to June, with a range from 10.0% (May, 1978) to 20.5% (November, 1976). An acetone fractionation scheme was developed with the main aim of concentrating these 2-oleo disaturated triglycerides into one fraction (the intermediate fraction) which may be useful as a cocoa butter replacer. The first precipitate (the hard fraction was separated by filtration, and the filtrate was adjusted for solvent : fat ratio and then further cooled to precipitate the intermediate fraction. After separation of this precipitate, acetone was distilled from the filtrate to produce a final fraction (the soft fraction). A screening experiment showed that the solvent : fat ratio at each crystallisation, the temperature to which the fat solution was cooled at each crystallisation, the water content of the acetone and the degree of agitation during crystallisation all affected the fractionation. The effect of these variables upon one sample of mutton tallow was studied, and mathematical models were developed to predict the yields of the three fractions and the melting properties of the intermediate fraction, The model predicting the melting properties of the intermediate fraction was used to estimate the fractionation conditions which would give an intermediate fraction with melting properties most similar to those of cocoa butter. From this, a fractionation was performed with first and second crystallisation temperatures of 9.2°C and 5.2°C respectively, solvent to fat ratios at the first and second crystallisations of 1.0:1 and 10.0:1 respectively, a water concentration in the acetone of 0.6% and a defined agitation condition. The yields of the hard, intermediate and soft fractions were 34.5 wt %, 2.5 wt % and 63.0 wt % of the tallow respectively. The intermediate fraction contained 51.0% of 2-oleo disaturated triglycerides (compared to 68.9% in cocoa butter) and had very similar melting properties to cocoa butter. Then the fractionation scheme was modified to give a greater yield of the intermediate fraction (8.3 wt %) but the melting properties of this intermediate fraction were less similar to those of cocoa butter. This latter fractionation scheme was scaled up (from 20 g tallow to 200 g and 1 kg). On each scale an intermediate fraction with consistent yield and melting properties was obtained. The yields of the other two fractions varied, however, and overall there was a considerable difference in the behaviour of the fractionations on each scale. Attempts on the 1 kg scale to produce an intermediate fraction with properties similar to those of the best 20 g intermediate fraction (i.e. similar to cocoa butter) were unsuccessful. The highest proportion of the important 2-oleo disaturated triglycerides attained in a 1 kg scale intermediate fraction was 36.8%, and this fraction melted, over a wider temperature range than cocoa butter. This intermediate fraction may be useful as a cocoa butter substitute in a coating chocolate, but is unlikely to be able to replace cocoa butter in chocolate. The hard fraction produced from this 1 kg fractionation (23.0 wt % of the tallow) showed promise in a baking shortening blend with butter, but was too hard to be useful as a pastry shortening. The soft fraction performed well as a deep-frying medium and in mayonnaise.