Commercial pasteurized, UHT and sterilized milk samples, and laboratory autoclaved milks, were analysed for lactulose by an enzymatic method. No lactulose was detected in the pasteurized milk. On the basis of the measured lactulose concentrations, the UHT and sterilized milk samples fell into five groups according to the type of process that they had received. The lactulose concentrations in the laboratory autoclaved milks rose on storage, but this was not observed in the other milks. It appears that pasteurized, UHT and sterilized milks can be differentiated by their lactulose content.

Formation of lactulosyl-lysine 666 UHT and sterilized milks Lactulose in heated milk 669 Methods of analysis 676 Mechanisms for its formation 669 Chromatographic methods 676 Clinical significance of the occurrence of 670 Spectrophotometric methods 677 lactulose in heated milks Enzymic determination 678 Baby foods 670 Conclusions 678 Enzymic hydrolysis 671 References 679

24 ultra heat treated (UHT) and 17 in-bottle sterilized milks prepared in commercial plants and of a similar age were presented to taste panels for assessment of flavour and colour. The colour was also analysed instrumentally and the lactulose content of the milks measured by an enzymic method. Flavour defects, other than the cooked flavour, were scored very low. The lactulose content of the milks was negatively correlated with the acceptability of the milks and positively correlated with the cooked flavour. Correlations were also found between the visual and instrumental assessments of milk colour and between some of the instrumental colour parameters and the lactulose content of the milks. UHT and in-container sterilized milks could not always be distinguished on the basis of their organoleptic qualities. The best distinction was observed in the scores for cooked smell and acceptability of taste. The usefulness of the lactulose content and/or the colour of the milks in discriminating between UHT and in-bottle sterilized milks was calculated. Using all the data, sterilized milk could be correctly identified in 96-7 % of decisions. The lactulose content of a milk was considered to give a good indication of the organoleptic quality of the milk.

Time/temperature profiles were obtained for commercial ultra high temperature (UHT) and sterilized milk processing plants and for a pilot UHT plant operating under nine different conditions. Samples of milk from each process were analysed for lactulose by an enzymic method, yielding concentrations of lactulose of 4 to 118 mg/100 ml. The measured lactulose concentration could be derived from the corresponding time/temperature profile by assuming an Arrhenius relationship with an activation energy of 152 k.J mol^1, for the whole range of UHT and in-container processes examined.Lactulose (4-0-/?-D-galactopyranosyl-D-fructose) is formed in heated milks by the alkaline epimerization of lactose catalysed by the free amino groups of casein (Adachi & Patton, 1961;Richards & Chandrasekhara, 1960). The determination of lactulose in heated milks has been shown to be useful in distinguishing pasteurized, ultra high temperature (UHT) and sterilized milks (Andrews, 1984).The rate of formation of lactulose in heated milk is both temperature and pH dependent. Ceicr & Klostermeyer (1983) autoclaved milks which had pH values between 6-59 and 6-72. They reported that milk at pH 6-59 contained 28% less lactulose than the control milk of pH 6 -70, and that milk at pH 6 -72 contained 9% more. However, they showed that in a commercial UHT process, when the raw milk was stored until the pH had fallen to pH 6-55 before processing, the processed milk contained only 8% less lactulose than when the fresh raw milk of pH 6 -7 was processed.In milk, pH and temperature are not independent variables: as the milk is heated, the pH falls. There are, therefore, two factors operating during heat processing that affect the rate of formation of lactulose.In order to relate a lactulose concentration to the heat process by which it was formed, it is necessary to determine the activation energy, E A . However, the effect of pH means that E A may not accurately describe the rate of formation of lactulose.To determine E A , the lactulose contents of milks processed in commercial and pilot plants were measured by an enzymic method. From the time/temperature profile of each process, the extent of the reaction was calculated for a range of values of E A . The E A value applying to lactulose formation in practice was the value producing * (irant aided by the Agricultural and Food Research Council.

Milk ultrafiltrate and milks of varying protein, citrate and phosphate concentrations were heated in sealed containers. Protein was found not to be involved in the mechanism of formation of lactulose, but increasing the protein content of milk reduced the concentration of lactulose after heating. This was considered to be due to increased condensation of lactose and lactulose with amino groups of the protein, Less lactulose was formed in milk ultrafiltrate than in skimmed milk accorded the same heat treatment, which was attributed to the buffering capacity of the milk protein in skimmed milk. Activation energies for lactulose formation in skimmed milk and in ultrafiltrate were 128 and 131 kJ/mol respectively. Citrate and phosphate catalysed the formation of lactulose. It is proposed that the formation of free lactulose in heated milk and ultrafiltrate proceeds exclusively by the Lobry de BruynAlberda van Ekenstein transformation with the naturally occurring phosphate and citrate acting as base catalysts.