Abstract
Factors affecting the membrane filtration of food suspensions were studied for 58 foods and 13 membrane filters. Lot number within a brand, pore size (0.45 or 0.8 micrometer), and time elapsed before filtration had little effect on filterability. Brand of membrane filter, flow direction, pressure differential, age (microbiological quality) of the food, duration of the blending process, temperature, and concentration of food in the suspension had significant and often predictable effects. Preparation of suspensions by Stomacher (relative to rotary blender) addition of surfactant (particularly at elevated temperature) and prior incubation with proteases sometimes had dramatic effects of filterability. In contrast to popular opinion, foods can be membrane filtered in quantities pertinent to the maximums used in conventional plating procedures. Removal of growth inhibitors and food debris is possible by using membrane filters. Lowering of the limits of detection of microorganisms by concentration on membrane filters can be considered feasible for many foods. The data are particularly relevant to the use of hydrophobic grid-membrane filters (which are capable of enumerating up to 9 X 10(4) organisms per filter) in instrumented methods of food microbiological analysis.