Tel.+49 30 838 62550 Fax.+49 30 838 46029
The error of colony counting consists of three components1. The unavoidable sampling error depending on the number of colonies / plate,2. Error, due to analysis - technical failures but also due to microbiological effects such as antagonism, synergism and overlapping,3. Error between samples, representing the dispersion between the elements of a batch.The significance of the three components of error have been tested on three individual batches of two kinds of vacuum - packed sliced sausage (Mortadella, well minced and Jagdwurst, coarsely minced). During 21 days of storage on 9 various days samples were taxen each consisting of three packages. Two dilution series Were made for each sample. Aliquots of each dilution were spread on three plates of ten select media using spatula technique.Based on the results the following can be concluded:1. All six batches had initially a similiar microbial profile. The lactobacilli became the prevailing flora during storage following a sigmoidal growth curve. Numbers of Brochothrix thermosphacta likewise increased continuously. Sorbic acid lactobacilli, enterobacteriaceae, coliforms occured in varying numbers but did not show a clear tendency of growth. The three production batches of Mortadella developped considerable differences of the microbial profile.2. The systematic error formed a smaller position of the total error than did the random caused error.3. The most important component of error was the error between samples. It sometimes reached 80 - 100 % thus exceeding the random caused error by a factor of twenty.4. Because of the marked error between samples results of colony counting were often affected by a high total error. The variation coefficient s / x (%) as referred to the median value of colonies of the three packages amounted 20 to 140 % with APC and up to 200 % with enterobacteriaceae and coliforms respectively.5. The influence of the variables of the experiments on the structure of errors is as follows:a. Kind of sausageThe systematic error of Mortadella was less striking than for the coarsely minced Jagdwurst. The error between samples did not allow any clear statement because of the structure of the experiment.b. Storage - periodThe number of colonies of the three packages were in the same range at the beginning of the experiment. The error between samples however increased the longer the period of storage lasted and the more the microorganisms multiplied.c. Number of colonies:There was a clear relationship between the numbers of colonies and the systematic error. Samples with colony forming units of less than 50 / plate did not reach even the sampling error (underdispersion), whereas a higher amount resulted in overdispersion.A systematic error was developped due to microbial antaggonism, synergism and overlapping but mostly its portion of total error was less than the unavoidable sampling error.d. Species of organism:The sampling error was increased by the total number ofaerobic colonies whereas the error between samples was increased by enterobacteriaceae, coliforms and Brochothrix thermosphacta.The considerable lack of precision of microbiological colony counting is caused by sampling error, systematic error between samples and results in a low representability of the data. The dispersion of the data hampers establishing of microbiological criteria. Accordingly, hygiene of production cannot be monitored by examining a single sample.