enteric bacteria in water involve filtration of 100 mL sample volumes through 0.45-μm-porosity, 47-mm-diameter nitrocellulose membranes and plating on selective media. Membrane filtration is commonly used for indicator microorganisms such as total coliforms, fecal coliforms, enterococci (fecal streptococci), and aerobic spore formers, and for detection of specific organisms or pathogens such as Escherichia coli, E. coli O157:H7, Shigella, Salmonella, Yersinia, and others. Membrane filtration of turbid surface water can be problematic and a labor-intensive Multiple Tube Fermentation method is often used. Detection of enteric protozoa and viruses requires analysis of much larger volumes of water than the 100 mL typically used for bacterial analysis because of their lower expected concentrations. In contrast to detection of enteric bacteria, conventional filtration approaches for protozoa and viruses require recovery of the organisms from the filtration matrix and further processing for analysis. Elution and dissolution of the filter matrix are commonly used. The selection of elution procedures and reagents is critical if cultural or infectivity assays are the end point analyses because some methods reduce the viability of the recovered organisms. Membrane dissolution methods often result in loss of infectivity, so these procedures are not suitable for assays in which viability or infectivity measurements are crucial.
Concentration methods for microorganisms in water have typically been optimized for a specific pathogen or at best for a limited number of related pathogens (i.e., the enteroviruses). Although there have been attempts to search for a single concentration method for an array of microorganisms in water, to date such efforts have produced mixed results. Development of new technology for concentrating pathogens in water is tedious and requires extensive testing with a variety of organisms and water matrices. Membrane concentration is perhaps the most explored concentration technique, and its advantages and disadvantages have been documented. Membrane concentration procedures are suitable for implementation in the field and for rapid throughput in the laboratory. Hollow-fiber and various ultrafiltration formats have been explored as sampling approaches for multiple bacteria, protozoa, and viruses in water (e.g., Sobsey et al., 1996).
Presnell and Andrews (1976) described a combined membrane filter-most probable number (MPN) procedure to increase the amount of water that could be passed (typically 4-5 L of surface water) through the membrane filter. The filters were subsequently washed and MPN analysis was conducted. Van Sluis and Yanko (1977) developed a concentration procedure at the County Sanitation Districts of Los Angeles County (CSDLAC) laboratory to detect the occurrence of Salmonella in disinfected effluents and receiving water. The procedure utilized Whatman glass-fiber filters overlaid with filter aid and a pressure filtration apparatus to concentrate 20-liter samples of surface water. The glass-fiber filter and filter aid were then emulsified in diluent. Salmonella were detected at concentra-