the water-supply op towns in the teopics 303 



Quantitative Examination 



The samples were plated on the usual nutrient agar (1 per cent, to phenol-phthalein) 

 and the quantities usually employed were '2 c.e., '3 c.o., '5 c.c. and 1 c.c. Occasionally, 

 2 c.c. plates were also poured. The plates were made in duplicate and often in triplicate. 

 As a control, one set of plates was poured by me from one water sample and another 

 set by my assistant from another sample. Now and then, the third sample was also utilised 

 for making plates. In addition, control plates were made with agar alone and with 

 agar to which sterile water was added. The plates were incubated for 48 hours at a Technique 

 uniform temperature of 37 ' C. and then counted. Occasionally, there was trouble with 

 the incubator, and when a temperature of 37° G. could not be maintained, the plates 

 were allowed to remain for a longer time in the incubator. 



As a rule, the technique proved satisfactory, but on one occasion it was found that 

 the stock agar had an inhibitive action on bacterial growth, owing, probably, to faulty 

 preparation or standardisation, and the counts made on this agar were discarded. 



Further, on several occasions the presence of haboubs, or high winds with dust, 

 resulted in air colonies getting access to the plates and ruining the work. It was found 

 impossible to prevent this contamination. 



Qualitative Examination 



The presence or absence of B. coli was determined at every examination. For this 

 purpose the following standard media were employed : — 



Glucose, bile-salt neutral-red broth with Durham's tube for gas formation, MacConkey's 

 glucose, peptone, bile-salt medium with Durham's tube, and latterly the new lactose-bile 

 medium of Jackson with Durham's tube, which is said to constitute the most delicate 

 presumptive test for the presence of B. coli* 



The quantities of water examined were as a rule 5 c.c, 10 c.c, 25 c.c, 50 c.e. 

 and 100 c.c. When any suspicious appearances were observed, confirmatory tests were 

 employed ; the organisms present being plated out on the Endo or Drigalski-Conradi 

 media or on lactose bile-salt neutral-red agar, and then examined for indol and subcultured 

 into the usual milk, sugar and litmus media as well as examined microscopically. 



I may say that, on the whole, the Endo fuchsin agar yielded the best results, savage's 

 When considered necessary, the B. enteritidis sporogenes test was applied and a search method 



closely 



made for streptococci. When essential, the pathogenic effect of a suspicious microbe foUowed 

 was tested on animals — gerbils, jerboas, fowls and guinea-pigs. In short, the method 

 favoured by Savage was closely followed. 



The condensation water in the air-pipe line, serving the air-lifts, was also examined 

 bacteriologically. 



Routine observations, as regards the temperature of the water for different wells tested, 

 and as regards the level of the ground water in the wells, were made while the tests 

 were being conducted. The data as regards levels were furnished by Mr. Williams — the 

 Engineer of the Water Works. 



I have said that the results of the first examinations were surprising. They were such 

 as to lead to an immediate condemnation of the water both on bacteriological and chemical 



* This has recently been confirmed by Hale and Melia, ^ * who also state that lactose-bile makes a distinction 

 between recent and distant contamination, and hence gives better evidence of the actual relative sanitary 

 quality of a water. Promme {loc. cit.), however, is not of this opinion. 



