The two species, _Bacillus typhosus_ and _B. coli_, agree in possessing the following characters: no spores, no liquefaction of gelatine; both grow well on phenolated gelatine, and in Parietti's broth; both act similarly upon animals, though typhoid fever is not a specific disease of animals. The _Bacillus typhosus_, though a somewhat susceptible bacillus, can when dried retain its vitality for weeks. In sewage it is very difficult indeed to detect, and is soon crowded out. Dr. Andrews and Mr. Parry Laws, in their bacterial researches into sewage for the London County Council,[15] found that when they examined specially infected typhoid sewage it was only with extreme difficulty they isolated Eberth's bacillus. In ordinary sewage it is clear such difficulty would be greatly enhanced. [Illustration: B. COLI COMMUNIS] We have pointed out elsewhere the relation between soil and typhoid. In water, even though we know it is a vehicle of the disease, the _Bacillus typhosus_ has been only very rarely detected. The difficulties in separating the bacillus from waters (like that at Maidstone, for example), which appear definitely to have been the vehicle of the disease, are manifold. To begin with, the enormous dilution must be borne in mind, a comparatively small amount of contamination being introduced into large quantities of water. Secondly, the huge group of the _B. coli_ species considerably complicates the issues, for it copiously accompanies the typhoid, and is always able to outgrow it. Further, we must bear in mind a point that is systematically neglected, namely, that the bacteriological examination of a water which is suspected of having conveyed the disease is from a variety of circumstances conducted too late to detect the causal bacteria. The incubation period of typhoid we may take at fourteen days. Let us suppose a town water supply is polluted with some typhoid excreta on the 1st of January. Until the 14th of January there may be no knowledge whatever of the state of affairs. Two or three days are required for notification of cases. Several more days elapse generally before bacteriological evidence is demanded. Hence arises the anomalous position of the bacteriologist who sets to work to examine a water suspected of typhoid pollution three weeks previously. There can be no doubt that these difficulties are very real ones. The solution to the problem will be found in Dr. Klein's dictum that "a water in which sewage organisms have been detected in large numbers should be regarded with suspicion"[16] as the vehicle of typhoid, even though no typhoid bacilli were discoverable. The chief of these sewage bacteria are believed to be _Proteus vulgaris_, _B. coli_, _P. zenkeri_, and _B. enteritidis_, and they are all nearly related to _B. typhosus_. The presence of the _B. coli_ in limited numbers is not sufficient to indicate sewage pollution, seeing that it is so widely distributed. But