COUNTING AND TESTING FOR MICRO-ORGANISMS 73 preparation, times and temperatures of incubation and culture media. Much useful data has been made available to control and pharmacopoeial authorities to enable them to consider the need for appropriate standards. In addition, the problem of defining precise official methods has been dis- played. Such detailed methods are considered to be essential in the event of disputes involving legislative requirements or acceptance or rejection of a shipment of product. This is of paramount importance with the rapid growth of international trade in pharmaceutical and cosmetic products as raw materials or in final forms. Whilst precisely detailed methods are desirable for this purpose, their availability and use will not necessarily mean that different laboratories will get the same results. For example, using standardized methods for the detection of Salmonellae in comparable and identical food samples, artificially contaminated in the laboratory, a collaborative study (46) produced results which varied widely. When more than 10 organisms were present in test samples fairly close agreement was reached by all laboratories but with smaller numbers of organisms there was disagreement. The differences were even more pronounced if organisms other than Salmonellae were also present in the samples. Such differences could occur in the testing of products for specific organisms as prescribed in the current editions of the U.S.P. (37) and N.F. (43) when there is a numerical superiority of other extraneous organisms occurring as natural flora in the product. In the performance of tests using selective media, the experienced eye is probably the most important tool used for diagnosis. Public health and industrial food laboratories have already acquired this expertise but such experience within pharmaceutical laboratories may be lacking for some time. Some of the procedures used in the investigations under discussion were relatively simple and non-selective in character. They could only serve as screening tests and could fail to detect or wrongly identify the specific organism being sought. All organismal growth on cetrimide agar cannot be labelled as Pseudomonas aeruginosa. Analytical procedures routinely used in clinical microbiology cannot always be indiscriminately applied to pharmaceutical products. The basic differences between products and clinical samples are liable to be overlooked. As a general rule, products, other than those of natural, vegetable or animal origin, have been shown to have a low incidence of organisms commonly classified as potential patho- gens compared with clinical samples. Unlike clinical samples, organisms in products may be subjected to physiologically debilitating processes such as heating, freezing, desiccation, irradiation, chemical action and extremes

74 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS of pH, temperature and osmotic pressure during manufacture and stor- age. There is also a possibility of interference with the selective performance of a particular testing medium when inoculated with a product or raw material which might create a biochemical imbalance in the system. This problem is particularly posed when using relatively large test samples (10-25 g) of product (37, 40, 43), the testing medium may then fail to detect the presence of the specific organism being sought. For example, inaccurate results would be obtained if methods of colimetry were to be indiscrimin- ately used for the detection of coliforms in a product containing sucrose as an ingredient of the formulation. Such a product would contribute signifi- cant amounts of sucrose to the lactose broth used in the presumptive test for coliforms and thus permit sucrose fermenting organisms to grow. Several products contain sodium chloride in significant amounts which could then act as a selective medium for Staphylococcus aureus by suppress- ing the growth of other important contaminants. Properly designed microbial tests must encompass the entire range of conditions which affect cellular metabolism. These conditions necessarily include growth enhancers, inactivators, appropriate temperatures and pH, and oxygen potential regulators. Just as it is vitally necessary to supply living organisms with nutrients required for growth and reproduction, so it is equally important to eliminate, as far as possible, those elements that inhibit growth and reproduction. In the investigations carried out little attention has been given to altera- tion of pH of the medium by the product. Adjustment of pH has been done in some cases (9) and is recommended by others (43), but in general no adjustment of pH has been carried out. This important aspect has either been ignored or else complete reliance has been placed tipon the buffering capacity of the diluting fluid or medium used. Since pH can influence the growth of organisms, much more attention is required to ensure that the pH of the medium used in counting and 'sterility test type' techniques is optimal so that growth of small numbers of organisms can take place. Numerous workers in this field have used methods for counting bacteria or fungi but few have given any details of the reproducibility of such counts. As has been experienced in other microbiological investigations, the reproducibility of counts is of a low order and they can vary with the exact type, or even the batch of medium used, the temperature of incubation and other factors. There are now several requirements by control and official compendia