IRREGULARITIES IN THE TEST FOR B. COLI IN WATER

The test for B. coli, while considered the most useful method available for determining the sanitary quality of water, is subject to certain irregularities, chief of which is the occurrence of pre-sumptive positive tests which fail to reveal the presence of B. coli group organisms when subsequently subjected to the usual confirmatory tests. These non-confirming presumptive positive tests, which occur with great frequency in certain waters, particularly after chlori-nation, have been ascribed by various investigators to the pres-ence of aerobic and anaerobic spore-forming lactose-fermenting bacteria and to gas-producing symbiotic groups. Several mediums, notably brilliant green bile, have been experimented with for the inhibition of these interfering organisms and exten-sive investigations of a similar nature are at present in progress. In Toronto, positive presumptive tubes which fail to confirm appear to be of at least two distinct types, (1) those probably due to one of the above causes, and (2) those in which colon group bacteria were originally present and lost in the confirmatory pro-cedure. The former, which occur more frequently and constitute over 90 per cent of the presumptive positives from the finally chlorinated water, while consuming a great deal of time and pro-longing the period required for interpretation of results, do not affect the reliability of the completed tests, whereas the second type, in which a negative result is recorded for a tube actually positive, may lead to errors of considerable magnitude. It is this latter type which will be discussed in the present paper. The following examples (table 1), selected from results actually obtained in routine examination of raw Lake Ontario wate