MICROBIOLOGICAL INTEGRITY OF COSMETICS 487 (example, Lotion A), changes in the in vitro test methodology were initiated. The chal- lenge inoculum was increased in mid-1980 to its current concentration of 106 cfu/gram of product for bacteria and 105 cfu of mold and yeast/gram of product. In addition, more stringent inoculum reduction requirements were instituted. Bacterial recoveries are required to be 10 cfu/g one week after each challenge inoculation and continue at that level throughout the test. Low initial recoverable counts of mold and yeast (0.1% of the original inoculum), which must reduce to 10 cfu/g by test completion, meet the criteria for passing. The selection of challenge organisms was changed to better simulate the population of organisms encountered during consumer use. Evaluations of organism resistance to standard cosmetic preservatives are frequently made. In addition, the accelerated preser- vation test was added as a more stringent challenge test for preservative efficacy. It is used to eliminate marginally preserved formulas earlier in the development cycle. Several additional modifications to the consumer-use testing program have taken place. Whenever failures occur, a chemical analysis of the preservative system in the used sample is compared to an unused sample from the same batch. This enables us to monitor preservative stability over time and after use. We currently isolate and identify organisms recovered from used products and, subsequently, use these contaminants to rechallenge the test product and to monitor its preservative resistance. This information allows a more thorough understanding of preservative efficacy during product use. The consumer-use panels for mascara evaluations now include testing of all new mascara formulations and any applicator or packaging changes. CONCLUSION The microbiological acceptability of a cosmetic is a function of its preservative efficacy during actual use conditions. The microbiologist strives to simulate consumer "chal- lenges" to the product in use by measuring the adequacy of the preservation system by means of an in vitro challenge test. The predictability of the in vitro test procedure is proven only through a well-controlled consumer-use testing program. Several modifica- tions to the in vitro test procedure have been made to increase this correlation and predictability. With all these parameters measured and validated accurately, the con- sumer is guaranteed a microbiologically safe cosmetic product. REFERENCES (1) A guideline for the determination of adequacy of preservation of cosmetics and toiletries, CTFA Technical Guidelines (1970). (2) Microbial tests, antimicrobial preservatives--Effectiveness, United States Pharmacopeia XXI, USP Pharmacopial Convention, Inc., 1151 (1984). (3) Efficacy of antimicrobial preservatives in pharmaceutical products, Appendix XVIC, British Pharma- copeia Volume II, Her Majesty's Stationery Office, London, A192-A194 (1980). (4) W. L. Bryon, E. R. Fizer, and J. K. Fartington, A review of methods for determining the preserva- tive efficacy of cosmetic products, Developments in Industrial Microbiology, 21, 273-276 (1980). (5) D. S. Orth, Establishing cosmetic preservative efficacy by use of D-values, J. Soc. Cosmet. Chem., 31, 165-172 (1980).

488 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS (6) M. Chan and H. Prince, A rapid screening test for ranking preservative efficacy, Drug and Cosmetic Industry, 34- 37 (December 1981). (7) J. J. O'Neill, C. A. Mead, and E. J. Scibienski, An accelerated preservation test. Presentation at SCC Annual Meeting NYC, 12/11/81. (8) R. Bhadauria and D. G. Ahearn, Loss of effectiveness of preservative systems of mascaras with age, Applied & Environmental Microbiology, 39, 665-667 (1980). (9) L. A. Wilson, A. J. Julian, and D. G. Ahearn, The survival and growth of microorganisms in mas- cara during use, American Journal of Opthalmology, 79, 596 (1975). (10) L. A. Wilson and D. G. Ahearn, Pseudomonas induced corneal ulcers associated with contaminated eye mascaras, American Journal of Opthalmology, 84, 112-119 (1977).