SUMMER DIARRHCEA l6l 



these animals, whereas of the other non -lactose fermenting 

 bacilli tested only 35% proved fatal. 



From the investigations, which have been quoted, it seems 

 evident that Morgan's bacillus is intimately associated with 

 summer diarrhoea in certain epidemics, whilst its connection 

 with other outbreaks has not been established. 



During 191 1 the writer (Graham-Smith, 1912, i.) carried out 

 the only extensive series of investigations on non-lactose 

 fermenting bacilli in flies which has yet been made. 



These observations were carried out for the purpose of 

 ascertaining the frequency and distribution in house-flies 

 (J/, domestica), obtained from different sources, of bacilli which 

 neither ferment lactose nor liquefy gelatin, and more particularl}^ 

 for the purpose of determining whether the varieties which occur 

 in districts where epidemic diarrhoea is prevalent differ from 

 those which occur elsewhere. A record was kept also of the 

 number of flies infected with lactose fermenting bacilli of the 

 colon type, and of the non-lactose fermenting bacilli which 

 liquefy gelatin. 



From 14th July to i6th October, 191 1, 1242 flies were 

 examined, of which 624 were obtained in Cambridge and 618 

 came from Birmingham. The Cambridge flies were collected in 

 several different localities ; 97 (Series B) from seven houses in 

 the town in which cases of epidemic diarrhoea had occurred, and 

 74 (Series C) from a farm house, situated on the outskirts of the 

 town, in which all the inhabitants suffered severely from diarrhoea. 

 The remaining flies were obtained from houses in which diarrhoea 

 did not occur, 157 (Series F) from a bake-house, 51 (Series E) 

 from a house in the centre of the town, 90 (Series H) from 

 a house outside the town and 155 (Series G) from a farm house 

 situated two miles out of the town. 



Dr Robertson, the Medical Officer of Health, very kindly 

 arranged for sending flies from Birmingham. Specimens were 

 received from 53 houses in which cases of epidemic diarrhoea 

 had occurred, and of these 471 (Series A) were examined. 

 Specimens were also received from seven houses, in the neigh- 

 bourhood of the others, in which no cases of epidemic diarrhoea 

 had occurred, and of these 147 (Series D) were examined. 



G.-S. 1 1 



