From the authors' private laboratory, Pretoria, South Africa CLOSE co-operation between laboratory and public health service leaves few cases of "enteric" undetected in Pretoria. In all cases a laboratory diagnosis is carried out and an effort made to trace the source of infection. For South Africa, our population (68,000 whites and 40,000 natives and coloured) is relatively stable, and continuity in methods and comparability of results are assured by the fact that for the past 16 years all typhoid laboratory investigations have been performed in the same laboratory. In this way it has, for instance, been possible to show that in Pretoria paratyphoid infections are practically non-existent-a circumstance which greatly facilitates work on carriers. In these 16 years we have dealt with more than 3000 cases of enteric, of which several hundred were diagnosed by means of blood culture. In only two instances were paratyphoid bacilli found, and in both cases the infection could be traced to sources outside the town. During the same period we have detected nearly 60 carriers, and in every one of these typhoid bacilli were found, and not a single strain of paratyphoid bacilli.In this paper we give the experiences collected over a number of years in tracing carriers, and also some more recent observations on Vi agglutinins in carriers.INFECTION CHIEFLY SPREAD BY CARRIERS Although we can trace but a relatively small number of cases of enteric directly to a carrier, we regard the carrier as the chief source of infection in our midst. Large explosive outbreaks, such as one associates with water supplies, have never occurred. We do have smaller "outbreaks ", sometimes more than one in a year, and these are practically always traced to a carrier in a dairy. There are very numerous small dairies, and this fortunately limits the extent of such outbreaks. Case to case infections occur, but the majority of our cases of enteric are sporadic, and do not facilitate the search for a carrier. It is not unusual to get a single case in a household, and a carrier amongst the native servants. Here, as in dairy outbreaks, one is impressed by the large number of persons who are exposed to infection and do not become ill. There must be many channels of infection, leading from chronic carriers through ambulatory cases and " silent infections " to new sporadic cases, channels which always remain hidden, and may incidentally give rise to new chronic carriers.Improved sanitation may, in the long run, put a stop to the carrier evil, both directly, and by decreasing the production of new carriers. But that time is not yet, and it is doubtful whether such measures will ever reach the whole