SUMMARY: Colony counts in 48 hr. at 37° were much higher in farm water supplies derived from rivers, streams, canals and unprotected wells than in water from springs, protected wells, upland surface waters and chlorinated supplies. Relatively few of the upland surface waters had counts exceeding 102/ml., but the counts for unprotected shallow wells were as high as those for rivers, streams and canals.
There was a general association between colony counts at 37° and presumptive coli‐aerogenes counts, but high 37° colony counts could not always be accepted as an indication of sewage or faecal pollution and obviously polluted waters sometimes had low colony counts at 37°.
Irrespective of the magnitude of the colony count, the majority (63%) of the colonies were Gram‐negative rods, though they formed a higher proportion of the flora in high than in low count waters. Aerobic spore‐forming rods constituted nearly 20% of the colonies and micrococci 10%, while Gram‐positive rods which did not form spores were present in small numbers at all colony count ranges. Streptococci were relatively rare, forming less than 2% of the colonies.
Approximately 25% of the 1,912 cultures isolated from colonies at 37° produced acid and gas in MacConkey's broth at 37°, but 7% of the 488 presumptive coli‐aerogenes cultures were aerobic spore‐forming rods. Typical coli‐aerogenes bacteria constituted 38% of the Gram‐negative rods and 24% of all the colonies examined. Surprisingly few colonies resembled Bact. coli type I in producing acid and gas in MacConkey's broth at 44°, and 13 (6.8%) of the 192 which did so were aerobic spore‐formers resembling Bacillus macerans.
Slightly over a third of the cultures produced an acid or proteolytic reaction in litmus milk within 2 days at 22°.
The results indicate that the colony count in 48 hr. at 37° does not contribute much useful information additional to that obtained from the presumptive coli‐aerogenes test at 37° and the Bact. coli test at 44°.