The potential for a revolution in microbial testing can be perceived with the near on-line detection of indicator microorganisms. By definition, these are microorganisms present in significant numbers within a food which, while not pathogenic, can be related through increasing count to the increased probability of pathogen contamination. We have used recombinant lux+ bacteriophage to detect enteric indicator bacteria without recovery or enrichment in 50 min, provided that they are present at levels greater than 10(4) g-1 or cm-2. After a 4-h enrichment, samples having enteric counts of 10 g-1 or cm-2 can be distinguished from background.

A study on the occurrence of Listeria monoeytogenes in. market samples of different food items indicated that 38% of the samples contained L. monocytogenes. The percentage of positive samples was highest among vegetables (49%), second in chicken (34%) and lowest in dairy products (26%). Inactivation of L. monocytogenes during cooking of green leaves indicated that L. monocytogenes was recovered from samples upto 8 min of cooking by which time the temperature was 72OC. L. monocytogenes strains isolated had haemolytic activity of 16 to 64 minimal haemolysis units by the microhaemolysis test.

The potential for a revolution in microbiol testing can be perceived with the near on‐line detection of indicator microorganisms. By definition, these are microorganisms present in significant numbers within a food which, while not pathogenic, can be related through increasing count to the increased probability of pathogen contamination. We have used recombinant lux+ bacteriophage to detect enteric indicator bacteria without recovery or enrichment in 50 min, provided that they are present at levels greater than 104 g−1 or cm−2. After a 4‐h enrichment, samples having enteric counts of 10 g−1 or cm−2 can be distinguished from background.

A total of 183 samples of consumable items of food and drink exposed for sale in the Kandy Municipal area were subjected to bacteriological examination with regard t o its hygienic quality. Only 19 per cent of the samples of ice cream examined were within the limits of the bacteriological standards laid down by the Sri Lanka Bureau of Standards. Other desserts (Fruit salad, Woodapple cream) show lower mean bacterial densities (Escherichia coli I, faecal streptococci and total viable count) than ice cream. The levels of faecal contamination of the fruit drinks did not vary with the place of preparation. This was due to the low pH. The faecal streptococci appears t o be better indicator of faecal contamination than E. coli fype 1 unticr freezing and at low pH.

The-objective of the study was t o compare the routinely used bacterial indicator systems of faecal pollution, the coliforms and faecal coliforms with faecal streptococci and spores of sulphite reducing anaerobes (Clostridia) t o assess their importance in determining the bacterial quality of water. A total of 8 4 samples of water from the Kandy clistribution system of water were tested. Percentage of samples that were found to be contaminated by the three indicators, coliforms, faecal coliforms and faecal streptococci are 37%. 15% and 54% respectively. I t was found that the use of faecal streptococci t o determine faecal pollution of treated tapwater has additional advantage over the presently used coliform, faecal colifom indicator systems. Results of the use of Clostndium spores in determining the hygienic quality of water was found to be unsatisfactory.