396 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS routine testing may not include microbial assay, and secondly even if it does, the examination of small batch samples may fail to reveal the potential dangers. Emulsions designed for topical application to the skin such as cosmetic creams and lotions need not, of course, be completely free from non- pathogenic bacteria and fungi, but the few organisms present in any product at the time of manufacture must be prevented from multiplying during the product's shelf and user life by an effective preservative. In general, preservatives are included in these products at a low concentration which is only sufficient to hold moderate numbers of organisms in a quiescent state, and an increase in preservative concentration is often undesirable because of increased cost and, perhaps more important, the increased risk of toxicity to the skin and mucous membranes. If, by any chance, unusually large numbers of organisms are introduced into a product during manufacture or packaging, a preservative which is normally perfectly adequate may be overcome with the result that the product is virtually unprotected against microbial spoilage. In these circumstances, the emulsion may break down because the emulsifier has been metabolized, off-odours may develop as a result of utilization of various constituents, gas may be produced, discolouration may occur if pigmenting bacteria are responsible, and perhaps worst of all because of its visibility to the customer, fungal growth may occur on the surface of the product and inside the pack. There are no hard and fast rules about the "best" types of preservatives for use in emulsions nor about the concentrations at which they should be used since all emulsions differ in their physical characteristics, and in the nature and concentration of their constituents. Some constituents will restrict microbial activity, while others will provide nourishment for growth some may be grossly contaminated themselves. The pH of the product may favour or discourage growth of some organisms, and the ratio of water to oil will also influence the likelihood of the survival and multiplication of micro-organisms. For all these reasons, a preservative must be selected on the basis of its suitability for one particular system, and not because it has proved effective in other systems. Suggestions about the preservation of emulsions are given by Wedderburn (1). Once a compatible and effective preservative has been selected, however, steps must be taken to ensure that it does not fail in the occasional batch of product when manufacturing conditions are not all that they should be.

HYGIENE IN MANUFACTURING PLANT AND EFFECT ON EMULSIONS 397 THE ORIGINS OF CONTAMINATION Water supplies, raw materials, sacks, bags, drums and vats in which ingredients are packed, air and dust, and the final product packing materials are all vehicles by which unusually large numbers of micro- organisms can enter even the best run factories. Unfortunately, spot- checks for microbial contamination do not provide a complete safeguard as pockets of organisms in otherwise satisfactory materials and equipment can often provide greater problems than moderate numbers evenly dis- tributed throughout. These pockets can easily be missed by the usual microbiological test techniques, and the only solution to this insidious problem is awareness of possible danger points, constant vigilance, and high standards of factory control. Demineralized water is contaminated more often than not, and the resin beds of ion exchangers are usually responsible. Baker (2) has cited a case where water entering an ion exchange unit contained four to five bacteria per m!, and left the apparatus carrying 500,000 per mi. Frequent sterili- zation of the resin beds in these units is desirable because the organic matter trapped in them provides excellent growth conditions for organisms. Eisman et a! (3) have also studied this problem and suggest that if the deionizing units are operated daily, contamination is less likely to occur than if they are only used intermittently and water is stored in large quantities where rapid multiplication can occur in short periods. Cruick- shank et al (4) found irrigation of ion exchange beds with 0.25 per cent formalin (0.1 per cent formaldehyde) satisfactory for sterilization. In factories where filtered tap water is used instead of demineralized water, the same attention should be paid to the filters as to deionizing beds. Frequent renewal of filters is, of course, preferable to sterilization particularly if organic matter collects in the filter, as this is often present in sufficient quantity to render the sterilizing fluid inactive against the organisms present. Stagnant water, and damp patches on walls, ceilings, and under stacks of raw materials are an eternal enemy to hygienic manufacture, as both bacteria and fungi will thrive in these conditions, become airborne, and eventually enter the plant. Equal, if less obvious, sources of aerial contamination are draughts which disturb dust and microbial spores. Of the raw materials entering a factory those of natural origin, such as the gums tragacanth, acacia and karaya, and solids like chalk, talc, kaolin and rice starch, are much more liable to carry large numbers of micro-organisms than the refined oils, fats and waxes, or synthetic materials,