WATER AND SEWAGE 701 
which develop in ordinary nutrient media at 20° C. and 37° C, a search 
for organisms characteristic of the excrement of man or animals, their 
approximate enumeration, and other tests which \'ary according to 
the source of the sample. 
Counting Bacteria. The counting of bacteria ordinarily signifies the 
numbers of microorganisms which will grow on gelatin incubated at 
20° C, and those that develop on agar at 37° C. Unpolluted waters 
usually contain relatively few bacteria that will grow at body tem- 
perature, consequentl\- the gelatin plate seeded with the same volume 
of water as the agar plate will show many more colonies than the 
latter; polluted waters show a more even distribution of types of 
bacteria that grow respectively at 20° C. and 37° (\ 
The amount of water to be plated in gelatin and in agar depends 
upon the source of the sample. Water from deep wtIIs and from 
springs should contain relatively few organisms, and a cubic centimeter 
of the sample is usually "planted." Surface waters almost invariably 
contain more bacteria than ground waters; it may be necessary to 
dilute a cubic centimeter of the sample with 99 cc. of sterile water to 
obtain the requisite distribution of organisms for an accurate estima- 
tion, or even higher dilutions may be necessary. Grossly polluted 
waters are diluted one thousand or even ten thousand times with 
sterile water before they are plated. In any event, not more than 200 
colonies or less than 50 colonies should be present in the final dilution, 
for experience has shown that greater muubers of organisms materially 
restrict development, and fewer than 50 colonies upon a plate intro- 
duces an error in dilution. 
Technique of Plating.— The sample of water, diluted to the required 
degree if necessary, is shaken vigorously to break up groups and chains 
of bacteria; 1 cc. of water is then removed with a sterile pipette into 
each of two sterile Petri dishes, being careful to prevent contamination. 
A tube of sterile nutrient gelatin (10 cc.) previously melted and 
cooled to 42° C, is then carefully poured over the water in one Petri 
dish, and melted nutrient agar is similarly poured into the other Petri 
dish. The water and culture fluid are intimateh' mixed by carefully 
tilting the plates, and then set aside to harden. The agar plate is 
in\'erted after it has hardened to prevent condensation of moisture 
up(m the surface of the medium; this procedure reduces the possi- 
bility of confluence of surface colonies. The gelatin plate is not 
iuAerted. 
Incubation at 20° C. for the gelatin plate and 37° C. for the agar 
follows. The agar plate is counted after forty-eight hours' incubation, 
the gelatin plate after four days. 
Interpretation of Bacterial Count.— At best the quantitative estima- 
tion of bacteria in water and sewage is inexact and relative only. 
The many factors of error in sampling, lack of uniformity in media, 
the difficulties of counting colonies when several hundred have grown 
in one plate— all tend to reduce the accuracy and precision of the 
