COUNTING AND TESTING FOR MICRO-ORGANISMS 71 material or product. In the investigations carried out, a variety of diluents have been used which appear to have been chosen in an arbitrary manner. For solids and liquids, the following diluents have been used or are recom- mended: nutrient broth (17, 18, 41) water (3, 37) water containing various surfactants (15, 41) phosphate buffer pH 7.2 (37) phosphate buffer and surfactant (24, 25) phosphate buffer pH 7.0, containing 0.93/0 sodium chloride (9) peptone water (38) quarter strength Ringer's solution (18). In a similar fashion the volume of diluent varies from 10 to 100 ml or more. For oily substances, creams and ointments, pre-treatments have been recommended involving mixing the sample with surfactant and then heating. The temperatures quoted (37, 40, 42) vary from 40 to 47øC and the times of heating also vary from being unspecified to 10 minutes. Whilst this procedure has much to commend it, it should be conducted with caution using minimal temperature and time. Some preservatives give inadequate protection to products and allow survival or growth of organisms at room temperature. If, however, the temperature of the product is elevated to 40-47øC for an unspecified or even short time, the preservative can exert a killing action and so reduce or eliminate the contaminating organisms. The result can therefore be misleading, especially if the product is contaminated with pseudomonads and also if the product contains quaternary ammonium compounds as either active ingredient or preserva- tive. TIME AND TEMPERATURE OF INCUBATION Prior to 1950, most sterility tests were carried out at 37øC for bacteria and 20-25øC for moulds and yeasts. In various compendia, the temperature was later reduced from 37øC to 32-35øC because it was thought that common airborne saprophytic bacteria presented a greater potential source of contamination of pharmaceutical products than the more fastidious pathogens. This temperature range was further reduced to 30-32øC following the unfortunate incident with con- taminated plasma in the U.S.A. when the pseudomonad contaminant responsible was killed at 35øC and was therefore not detected in the sterility test. In 1963, Pittman (33) recommended that there was no apparent reason why sterility tests for bacteria should be incubated above 32øC. The U.S.P. 18th rev. (37) has now reverted to a temperature range of 30-35øC. In the investigations reported, workers have used 30øC, 30-32øC, 30-35øC and 37øC for most of the counting and 'sterility test type' methods.

72 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Special temperatures of 42øC and 44øC have been used for Pseudomonas aeruginosa (22) and E. coil (17) in confirmatory tests for these specific organisms. The times of incubation used varies from less than 24 h to 14 days. CULTURE MEDIA Many media are being used for conventional sterility testing and the formulae for these appear in official compendia of many countries. A World Health Organization Study Group Report (32) listed nine media for culturing bacteria and six for culturing fungi. The Group could not recom- mend any one media in preference to another because of lack of compara- tive data. Since no single medium will support the growth of all bacteria, moulds and yeasts, more than one medium should be used. The question of which to use has been, and still is, the subject of many conferences and published reports. Pittman (33) concluded that none of the reports have supplied adequate support for the selection of any medium in preference to others in use. A possible shortcoming of published work is that the growth promoting properties of various media were determined by using micro- organisms considered to be likely contaminants, rather than using organ- isms actually isolated from contaminated products. She advocated that emphasis should be given to the recovery of organisms subject to insult by preservatives or active ingredients in a product. Many investigators oppose the use of a selective medium in conven- tional sterility testing since the purpose of the test is to detect as many micro-organisms as possible, in both type and number. The U.S.P. (37) and N.F. (43) have replaced Sabouraud medium of pH 5.7 with a soy-bean- casein digest medium of pH 7.3 because of its ability to detect bacteria as well as fungi. Bowman et al (35) have produced evidence from a collaborative study of twelve laboratories that soy-bean-casein digest medium is superior to Sabouraud medium for sterility testing by the mem- brane filtration technique. DISCUSSION As stated previously, the methods used for the enumeration and detec- tion of micro-organisms in products consist basically of counting and 'sterility test type' techniques for specific organisms. Many variations of these test methods have been employed together with variations in sample