Microbial source tracking is necessary because standard methods of measuring fecal contamination in water by enumerating fecal indicator bacteria (FIB) do not identify the source or sources of the contamination. Source tracking methods can be divided according to whether or not they require a library (a "host origin database" or set of bacterial isolates from fecal samples of known origin). Methods can also be divided by whether or not they use FIB and by whether or not they require cultivation of microbes. These variables all affect the cost, time required, and effectiveness of the methods. Prominent library-​dependent methods include antibiotic resistance analysis (phenotypic) and fingerprinting analyses (genotypic) such as ribotyping, REP-PCR and PFGE. Library-depend methods are costly, time consuming, and have complex and sometimes poorly understood requirements for sample sizes and analyses. In addition, new libraries are needed for each geographical region. Prominent non-​library-​dependent methods include chemical markers (e.g. fecal sterols/stanols​, caffeine and laundry whiteners) and host-specific PCR (e.g. of Bacteroidales molecular markers). Another set of techniques involves direct molecular monitoring of human viruses such as adenoviruses and enteroviruses. Viruses are important because they are not well correlated with FIB but are important pathogens. All of these methods have been tested at the proof-​of-​concept level, but there have been few organized comparisons and proficiency tests with blind samples. In one such study, the SCCWRP study, host-specific PCR performed well, as did ribotyping and PFGE. Other comparative studies have found somewhat different results. All support the following conclusions: 1) No method is quantitative; 2) Results from the same method differ depending on the operator and on differences in experimental design and analysis. Few studies have followed up the results of fecal source tracking to quantify resulting gains in water quality. The best evidence supports taking a multi-tiered approach to source tracking, moving from general to specific and from less to more expensive. The first step is intensive surveys using FIB, to target sources spatially and temporally. Once "hot spots" are identified, then very directed source tracking can be done if needed, starting with less-expensive methods that identify human contamination, and continuing to more-expensive ones as needed, to identify common species, or finally to identify all species. Companies that offer source-tracking services should be provided with blind proficiency samples to assess their abilities and estimate possible benefits, before they are hired. Water quality standards were established based on epidemiological studies that measured human health outcomes following recreational exposure to human-derived fecal contamination. There are no similar studies for exposure to animal fecal contamination, although it is logical to assume that the risk from animal fecal contamination is lower. Thus even if microbial source tracking shows that fecal contamination is animal-derived, there is usually no way to allow for a higher permitted level of FIB. Hence the benefits from microbial source tracking at the present time are only that it allows the source or sources of fecal contamination to be accurately assigned, located, and corrected. In some cases this could lead to a reduction in FIB. In other cases where the source is primarily wildlife and there is no way to control the wildlife, no immediate water quality benefit from microbial source tracking will be seen. National environmental health agencies must take the responsibility to fund the required epidemiological studies so microbial source tracking can be properly applied to estimate human health risk.