Ultra-high-temperature (UHT) processing of milk involves heating the product to 137-145 °C for a holding time of 8-2 s. This brings about a substantial difference in flavour between UHT milk and the more commonly consumed pasteurised milk (which is heated to 72 °C for 15 s). The aims of this project were to: determine the extent to which the flavour difference affects UHT milk consumption in Australia; develop a reliable objective method of measuring the stale flavour volatiles in UHT milk; to successfully apply this method to the analysis of stale flavour volatiles in UHT during storage; and devise a way of modifying the flavour of UHT milk to suit the consumer. A telephone survey was conducted in Melbourne and Brisbane to obtain a profile of milk consumption in Australia and to determine consumers' attitudes regarding UHT milk. It was anticipated that this survey would reveal the reasons for the low level of UHT milk consumption in Australia. Pasteurised milk was the main milk type used by more than 80% of respondents. For UHT milk this figure was much lower (approximately 10%), even though two thirds of respondents had tried UHT milk. Factors that were found to influence UHT milk consumption included existing milk consumption habits, consumer perception, flavour and price. The majority of nonusers of UHT milk stated "habit of using other milk type" as their main reason for not using UHT milk. Other reasons included "poor nutritional value", "poor flavour" and "not real/pure milk", indicating a negative consumer perception of the product. The flavour of UHT milk was identified as a problem, with nearly half of UHT milk users considering it to be worse than the flavour of pasteurised milk. However, a small proportion of UHT milk users preferred the flavour of UHT milk, with the majority of them stating that it was creamier, richer and/or stronger than the flavour of pasteurised milk. Prior to post-farmgate deregulation, price was shown to discourage consumers from using UHT milk. At the time of the survey, post-farmgate prices in Victoria were deregulated resulting in UHT milk being priced below that of pasteurised milk in some instances. This was believed to have contributed to a significantly higher market share of the product in Melbourne than in Brisbane. Solid phase microextraction (SPME) offers a solvent-free and less labour intensive alternative to traditional flavour isolation techniques. In this instance, SPME in conjunction with GC-FID was optimised for the extraction of 17 stale flavour volatiles (C3-11,13 methyl ketones and C4-10 saturated aldehydes) from the headspace of full-cream UHT milk. A comparison of relative extraction efficiencies was made using three fibre coatings, three extraction times and three extraction temperatures. Linearity of calibration curves, limits of detection and repeatability (coefficients of variation) were also used in determining the optimum extraction conditions. A 2 cm fibre coating of 50/30 µm divinylbenzene/Carboxen/ polydimethylsiloxane in conjunction with a 15 min extraction at 40 °C were chosen as the final optimum conditions. This method was used as an objective tool for monitoring the flavour quality of commercial UHT milk during storage. Methyl ketones, aldehydes and free saturated fatty acids were measured in the headspace of samples of two indirectly processed and two directly processed commercial UHT milks during room temperature storage for 16 weeks. All methyl ketones and aldehydes increased during storage, with fi-ee saturated fatty acids exhibiting little change. On average, the total methyl ketone and aldehyde concentrations in the indirectly processed UHT milks were higher than those in the directly processed samples. A strong correlation was found between the concentration of methyl ketones and various heat indices (furosine, lactulose and undenatured whey proteins) in the milk samples. In the final experiment, indirectly processed UHT milk was packaged aseptically into 2 1 laminated pouches with (treatment) or without (control) oxygen-scavenging film and stored for 14 weeks at 26 ± 0.3 °C. Samples were analysed at 0, 4, 8 and 14 weeks for dissolved oxygen, stale flavour volatiles (methyl ketones and aldehydes) and free fatty acids. Discriminative subjective analysis of odour by a consumer panel was also conducted. The oxygen scavenging film was shown to significantly (P<0.05) reduce dissolved oxygen content by 23-28% during storage. Significant (P<0.05) reductions of 23-41% were also observed for some stale flavour volatiles, including five methyl ketones and two aldehydes. Free fatty acid levels remained far below threshold values, indicating that lipolytic rancidity would not interfere with the subjective analysis. However, no significant difference in odour was detected.