ASSESSMENT OF THE PRESERVATIVE CAPACITY OF SHAMPOOS 237 chain. Hydrolysis of the sulphate groups leaves a fatty alcohol, in this case lauryl alcohol. It is suggested that the oxidation of this residue proceeds via lauryl aldehyde to lauric acid, and its subsequent [•-oxidation consists of the sequential removal of C,. units. Analysis of growth media by glc for intermediates of lauric acid breakdown showed that degradation did in fact occur in C2 units. Alternatives to this procedure include co-oxidation, i.e. oxidation starting at the distal end of the carbon chain, either before, after or concomitant with sulphatase activity. co-Oxidation must also occur since it is known that bacteria grown on nutrient broth alone are not able to hydrolyse the sulphate group but can utilize the detergent as their sole carbon and energy source. Therefore, sulphatase enzymes are necessary only when the detergent is present as the sole sulphur source. Pretreatment of the inoculum with low levels of the corresponding detergent to ensure fully induced enzyme systems failed to produce organ- isms with increased ability to survive in a shampoo. Organisms grown in nutrient broth, which is more convenient and certainly gives better yields, exhibited the same pattern as those grown in detergent solutions. Therefore, the induction of enzymes which allow utilization of shampoo detergent may not be of over-riding importance prior to the challenge test. It is not the reason for the delayed recovery of organisms after entering the shampoo. Pretreatment of the organisms with levels of detergent up to 13•o did, however, give the inoculum added resistance and allowed a rapid establish- ment of the population in the product. This is probably related to the slime covering the bacteria of the inoculum. Organisms present in shampoo produce slime in abundance and often display 'strings' of slime through the product, which suggests that the slime may be present as a coat either pro- tecting the cell membranes from high detergent concentrations, affecting induction of a permease or elevating induced enzyme levels. We have shown that growth of bacteria in high levels of detergent causes adaptation to that particular detergent. This was not related to the levels of sulphatase enzymes since bacteria with fully induced sulphatase still die upon entry into a shampoo, but may well be related to the type of slime coat formed. After growth in nutrient broth the bacteria had the ability to survive in different shampoos. Nutrient broth, therefore, remains a suitable growth medium for these test organisms. Unpublished evidence from this laboratory suggests that discrepancies between laboratory challenge tests and the actual in-factory situation regarding the preservative capacity of a product can arise not only because of changes in the test bacteria, but also because of differences between

238 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS batches of raw materials, especially detergents. Three separate samples of SLES provided by the same supplier gave three different results in a challenge test. Further investigation showed that one contained mercurials, one formalin and the other no preservative. It is clear that a challenge test on a product containing the preserved detergent would indicate that the formula- tion was self-preserving. The same formulation, prepared in the factory with unpreserved detergent, would soon become contaminated, and the laboratory challenge test would fall into disrepute. It is essential that the bacteria used in a microbiological challenge test for shampoos are given every opportunity to survive in the absence of a preservative so that the preservative system added subsequently is really put to the test. This being the case, the most important aspect of a challenge test is the choice bf the correct organisms: tap water provides a supply. This removes the need for maintaining bacteria on enriched media which could bring about metabolic or morphological changes, and it simulates actual contamination as it can occur in the factory. Special pretreatment of the bacteria before inoculating into a shampoo is not essential, except perhaps training them to withstand high levels of detergents. With some variables still remaining, the challenge test can still only give a guide to the likelihood of contamination in a product, though a much better guide than before. At a time when shampoos are containing more nutrients, allowing the growth of a much larger variety of organisms than has been found to date, it is essential that the most appropriate and informa- tive test methods be used. (Received: 24th January 1972) REFERENCES (1) Hsu, Y. Detergent-splitting enzyme from t'seudomonas. Nature, Lond. 207 385 (1965). (2) Taylor, E. W. Metropolitan Water Board 43rd Report on the Results of the Bacteriological, Chemical and Biological Examination of the London Water for the Years 1967/1968. (3) Yasufuku, M., Hashimoto, Iq ., Hamai, J. and Uesugi, Y. A biological study of t'seudomonas contaminant in shampoo and its chemical preservation. 5th Cong. Int. Fed. $oc. Cosmet. Chem. (Tokyo, 1968) Reprints 363 (1968).