Public health protection requires an indicator of fecal pollution. It is no
t necessary to analyse drinking water for all pathogens. Escherichia coli i
s found in all mammal faeces at concentrations of 10 log 9(-1), but it does
not multiply appreciably in the environment. In the 1890s, it was chosen a
s the biological indicator of water treatment safety. Because of method def
iciencies, E. coli surrogates such as the 'fecal coliform' and total colifo
rms tests were developed and became part of drinking water regulations. Wit
h the advent of the Defined Substrate Technology in the late 1980s, it beca
me possible to analyse drinking water directly for E. coli (and, simultaneo
usly, total coliforms) inexpensively and simply. Accordingly, E. coli was r
e-inserted in the drinking water regulations. E. roll survives in drinking
water for between 4 and 12 weeks, depending on environmental conditions (te
mperature, microflora, etc.). Bacteria and viruses are approximately equall
y oxidant-sensitive, but parasites are less so. Under the conditions in dis
tribution systems, E. coli will be much more long-lived. Therefore, under m
ost circumstances it is possible to design a monitoring program that permit
s public health protection at a modest cost. Drinking water regulations cur
rently require infrequent monitoring which may nor adequately detect interm
ittent contamination events; however, it is cost-effective to markedly incr
ease testing with E. coli to better protect the public's health. Comparison
with other practical candidate fecal indicators shows that E. coli is far
superior overall.