650 BACTERIA IN WATER. 



water-supply in small numbers as compared with the number of saprophytic 

 bacteria, it is not an easy matter to demonstrate the fact by the ordinary 

 plate method, especially in the case of non-liquefying species like the typhoid 

 bacillus. If we have, for example, one typhoid bacillus to one thousand ba- 

 cilli of other species it is evident that in a series of three plates, made in the 

 usual way for the purpose of obtaining isolated colonies, there would be but a 

 small chance of obtaining a colony of the typhoid bacillus in plate No. 3, 

 and a plate containing one thousand colonies or more would be so crowded 

 that the detection of the single typhoid colony would be very difficult. For 

 this reason, it is necessary to resort to special methods by which the more 

 numerous saprophytic bacteria will be excluded, or their numbers greatly 

 reduced. Some of the methods which have been successfully employed for 

 the detection of the typhoid bacillus and of the cholera spirillum are given 

 in the sections devoted to these microorganisms. We give below some de- 

 tails relating to the methods employed by bacteriologists of recognized com- 

 petence in recent investigations : 



Marpmann (1895) considers all water which contains faecal bacteria as 

 dangerous as a supply for drinking purposes. For the detection of patho- 

 genic bacteria he recommends the following procedure : 



The pathogenic bacteria are divided into two groups by cultivation in nu- 

 trient agar containing 0.2 percent of citric acid, and in the same medium con- 

 taining two per cent of sodium carbonate. The bacilli of the typhoid group 

 are said to grow in the acid medium but not in that containing two per cent 

 of sodium carbonate. On the other hand, cholera vibrios develop in the al- 

 kaline medium but not in that containing 0.2 per cent of citric acid. The ba- 

 cilli of the colon group also (" cloaca-bacilli") do not grow in the medium 

 containing citric acid. Bouillon containing the same amounts of acid and 

 alkali is also employed. The water to be examined is first mixed with an 

 equal portion of acid and of alkaline bouillon in two test tubes, and these are 

 kept at a temperature of 30 C. for twenty -four hours, during which time 

 the pathogenic bacteria, if present, will multiply and cause a clouding of the 

 culture media. Inoculations are now made into the acid and alkaline agar 

 and gelatin. Growth in alkaline gelatin at the room temperature (10 to 18 

 C.) is due to "cloaca-bacteria" ; growth in acid gelatin at 20 to 23 C. is due 

 to bacilli of the typhoid group. Plates should also be made from the clouded 

 bouillon, acid and alkaline ; and the colonies resembling those of the typhoid 

 or of the colon group should be tested in nutrient gelatin containing sugar 

 to ascertain whether there is development of gas, in which case the bacilli are 

 of the colon group. 



When typhoid and colon bacilli are associated in water the last-mentioned 

 bacillus takes the precedence, and the typhoid bacillus has a tendency to dis- 

 appear. This is shown by the experiments of Girnbert (1894), who introduced, 

 at the same time, colon bacilli and typhoid bacilli into water, and found that 

 at the end of forty-eight hours he was no longer able to isolate the typhoid 

 bacillus from plates. In view of this fact failure to find the typhoid bacillus 

 does not relieve the water from the suspicion of being dangerous if the colon 

 bacillus is present. But, on the other nand, this bacillus is so common that 

 it is perhaps the exception when it is not present in surface waters. As 

 pointed out by von Freudenreich (1895) it may, however, escape detection 

 unless a considerable quantity of water is used in making the test. When 

 the quantity is from one hundred to five hundred cubic centimetres, in- 

 stead of from one to five cubic centimetres, as was formerly the usual amount 

 employed, it is found not infrequently even in spring water (von Freuden- 

 reich). 



The author last mentioned says that when present in small numbers it 

 may be demonstrated by the method of Vincent, as follows : Mix of the water 

 ninety cubic centimetres with ten cubic centimetres of a twenty-per-cent 

 solution of peptone, and one cubic centimetre of a seven-per-cent solution of 

 carbolic acid ; place in the incubating oven at 42 C. If development oc- 



