PROCEEDINGS OF SECTION I. 591 



escape into it of raw faecal matter or to a more continued entry 

 of infectious material. Explosive outbreaks also occur in other 

 circumstances, and there are notable cases after the common eating 

 of a certain food-stuff as by diners at a banquet. 



Au unrepeated explosive outbreak is rather characteristic of 

 such a mode of infection, but is quite possible under several other 

 circumstances such, for instance, as a milk infection where the 

 combination of circumstances permitting the infection of the milk 

 has only occurred once. 



The discovery of chronic carriers of the typhoid bacillus has 

 undoubtedly thrown much light on the origin of certain outbreaks, 

 but has in many ways complicated the question. Nowadays one 

 has always to remember that the original source of the infective 

 excreta may be either a case or a carrier, and also that the carrier 

 may give no history of typhoid fever, and present no symptoms 

 of illness, and may show no evidence of antibodies in the blood, 

 and give a negative Widal reaction (9). 



I do not propose, in this necessarily sketchy paper, to deal at 

 any length with the carrier problem, but would here give a word 

 of warning to " carrier " enthusiasts. A carrier is, at most, a 

 source of infection, no doubt often a very puzzling element, be- 

 cause of his or her special possibilities of spreading the bacillus 

 possibly over a large area, whereas the range of action of a case 

 is necessarily limited, but, provided we keep it in mind that the 

 origin of the spread may be the faeces of either case or carrier I 

 think we will do well not to allow the carrier idea to dominate 

 our minds. It is well to remember that there are numerous 

 well-attested instances of other modes of infection, and there seems 

 a danger nowadays, in the hurry to track down carriers, of for- 

 getting other, shall I say more " old fashioned," ideas. It is 

 interesting to note how seldom nowadays one hears of water in- 

 fection, and yet there are the seemingly incontestable bacterio- 

 logical facts shown in the records by Losener (10), Kubler and 

 Neufeld (11), Hankin (12), Fischer and Flatau (13), Tavel (14), 

 Konradi (15), Jaksch and Rau (16), Snolsener (17), and others 

 to prove that it certainly is a big factor to be reckoned with. 



Bacteriological work does not properly come into use in this 

 part of the investigation of an epidemic until, from a considera- 

 tion of the facts obtainable by the above-mentioned methods, the 

 evidence begins to point in one or other direction. But let me 

 not be misunderstood — the bacteriologist is all the time being 

 employed in the early diagnosis of cases as they crop up, and is thus 

 materially aiding in the work, as each case as soon as diagnosed 



