HYGIENIC MANUFACTURE AND PRESERVATION 753 added to the bottles before sterilization as a 3% solution to give a final concentration of 18 mg 1-1 in the water sample, the amount depending on the size of the bottle, e.g. 0.1 ml is required for a 170 ml bottle. Sodium thiosulphate (Na2S2035H20) at a concentration of 18 mg 1-1 has no significant effect on the coliform organisms and should neutralize up to 5 mg 1-• of residual chlorine. 1.52 Colony count Water may contain a variety of micro-organisms having different optimum temperatures of growth. Most bacteria capable of growth in water will, in laboratory media, grow better at 22 ø than at higher tempera- tures. Organisms which grow best at 37 ø usually grow less readily in water and are more likely to have gained access from external sources. Since these two groups of organisms differ in their significance, it is desirable to count them separately. For this reason two sets of plate counts are usually prepared, one of which is incubated at 20-22 ø and the colonies counted after three days, the other of which is incubated at 37 ø , the count being made after 24 h. The number of dilutions used may be varied in the light of experience with waters of which the bacterial content is known. Normally original sample and dilutions of 10-• and 10-2 are usually sufficient for this purpose. (Appendix B, Section 1.215 p. 746 for method of plating out). Details of special procedures relating to sampling, the membrane filter, the identification of E.coli (fermentation of lactose at 44 ø) and other appropriate biochemical tests, also special methods for the isolation of pathogens are described in the official report (1). 1.53 Reporting plate count results Report the colony count as the number of colonies developing per ml of the original water on Plate Count Agar after 24 h at 37 ø or 3 days at 20-22 ø as the case may be. REFERENCE (1) The bacteriological examination of water supplies. Report No. 71 (1969). (H.M. Stationery Office, London). 1.6 Microbiological status of atmospheric environment A high level of microbial contamination in the air of a cosmetic manu- facturing plant is obviously undesirable as it may lead to contamination of the product. Aerial counts of bacteria, moulds and yeasts are liable to fluctuate widely in relatively short periods of time, e.g. due to air currents, opening of packing cases and the movement of personnel. Respiratory

754 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS infections affecting staff may well lead to localized high levels of atmos- pheric contamination. Apart from the widely varying levels of atmospheric contamination actually present, assay techniques may lead to extremely variable findings depending, for example, on whether they record static or dynamic condi- tions of the air being sampled. For these various reasons, quantitative standards cannot be recommended for general application. It is, however, suggested that atmospheric contamination should be monitored along with other possible sources of contamination, in order to gain a general picture of the microbiological status of a production unit. Suggested methods for use with a slit sampler* are impingement on selective media or membrane filtration after impingement in broth. A suitable range of media would be Nutrient Agar and MacConkey Agar for the isolation of bacteria, Sabouraud and Malt Extract Agar for the isolation of yeasts and fungi. 9,. PRODUCT DEVELOPMENT 9,.11 Assessing preservative capacity Whatever the preliminary methods of evaluating the effectiveness of a preservative, the final test should always be carried out on the complete formulation in the final pack. Most procedures are based on a dynamic or spoilage type of test in which the sample to be tested is challenged by inoculation with a selection of appropriate spoilage organisms. The con- taminated sample is subsequently examined for evidence of microbial activity during a test period usually lasting several weeks or preferably, in the case of moulds, for several months. If micro-organisms fail to grow in such a test, the sample can be re-challenged to provide further information on the stability of the preservative system. 2.12 Challenge organisms The range of micro-organisms examined in preservation tests is logically selected from those usually responsible for spoilage and others associated xvith common infections. Micro-organisms used in preservation studies must be vigorous strains and should include recent isolates from contaminated samples. Along with locally-isolated laboratory and factory contaminants, *A suitable piece of equipment for testing air samples under standard conditions is Airborne Bacteria Sampler Mark II. From: C. F. Casella and Co. Ltd., Regent House, Britannia Walk, London, N. 1.