coliforms and faecal streptococci to that found in humans, the relatively higher ratio of C. perfringens spores found in dog faeces may be a useful indicator when fresh faecal contamination is being investigated (Leeming et al., 1998).
It is important to note that spores of C. perfringens do not act as an indicator for non- sewage or animal faecal contamination in general, and therefore they are only suitable as indicator organisms for parasitic protozoa and viruses from sewage-impacted waters (Payment and Franco, 1993; Ferguson et al., 1996). Their resistance to disinfectants may also be an advantage for indexing disinfectant-resistant pathogens. Simple anaerobic culture is possible for C. perfringens spores after a short heat treatment to remove vegetative cells. Confirmation of their presence may be assisted by the addition of a methylumbelliferyl phosphate substrate to the growth medium (Davies et al., 1995).
Other indicator organisms for sewage, but also specific for human sewage are the bacteriophages to Bacteroides fragilis HSP40. These B. fragilis phages appear to survive in a manner that is similar to the hardier human enteric viruses under a range of conditions (Jofre et al., 1995; Lucena et al., 1996). Their numbers in sewage such as the F-specific RNA bacteriophages may be an order of magnitude lower than various coliphages. Furthermore, only 1-5 per cent of humans may excrete these phages (Leeming et al., 1998), and thus they may be unsuitable pathogen indicator organisms for small communities. The International Office for Standardization (ISO) standard methods for these phages are under final review (ISO, 1999c).
The ratio between thermotolerant coliforms and faecal streptococci has been proposed by Geldreich (1976) as a means of distinguishing between human and animal-derived faecal matter. However, this method is no longer recommended (Howell et al., 1995) and none of the currently-used bacterial indicators distinguish different sources of faecal matter confidently when used alone (Cabelli et al., 1983), although genetic typing of E. coli shows some potential (Muhldorfer et al., 1996). Identification of human enteric viruses can identify specifically the presence of human faecal material although the necessary procedures are difficult and expensive, and not readily quantifiable. Other microbiological options include specific identification of phenotypes of Bifidobacterium spp. (Gavini et al., 1991), Bacteroides spp. (Kreader, 1995), serotypes of F-specific RNA bacteriophages (Osawa et al., 1981) or, as previously discussed, the bacteriophages to Bacteroides fragilis (Puig et al., 1997). However none of these organisms are suitable for quantifying the proportion of human faecal contamination. Moreover, no one indicator or single approach is likely to represent all the facets and issues associated with faecal contamination of waters.
Recently, Leeming et al. (1994, 1996) demonstrated the ability to distinguish human from herbivore-derived faecal matter using a range of faecal sterol biomarkers (Table 8.1). The distribution of sterols found in faeces, and hence their source-specificity, is caused by a combination of diet, the animal's ability to synthesise its own sterols and the intestinal microbiota in the digestive tract. The combination of these factors determines “the sterol fingerprint”. The principal human faecal sterol is coprostanol (5β(H)- cholestan-3β-ol), which constitutes about 60 per cent of the total sterols found in human faeces. The C29 homologue of coprostanol is 24-ethylcoprostanol (24-ethyl-5β(H)- cholestan-3β-ol). In large quantities (relative to coprostanol), this faecal sterol is indicative of faecal contamination from herbivores. It is possible to determine the contribution of faecal matter from these two sources relative to each other by calculating