434 BACTEBIOLOGY. 



Inasmuch as the bacteria are liable to multiply rapidly, especial- 

 ly if the water is taken from a cool source and is then kept at 

 ordinary temperature, it is advisable to plate the water at the time 

 it is collected. Under these circumstances, instead of plates or Petri 

 dishes, flat flasks or bottles can be used. These contain the requisite 

 amount of sterile gelatin which is inoculated with the water as soon 

 as it is drawn. The flask is placed on its side, and, when the gelatin 

 solidifies, it can be taken back to the laboratory. 



If only a small number of colonies are present they 

 can be counted with the unaided eye, but when, as it fre- 

 quently happens, the number is very large, it is desirable 

 to make use of a counting apparatus. Fig. 57 shows the 

 counting apparatus of WolfEhiigel, which is usually em- 

 ployed when ordinary glass plates are used. The gelatin 

 plate, on which the colonies are to be counted, is placed on 

 the black glass base and covered with a glass plate ruled 

 into squares. The number of colonies under each square 

 can thus be easily determined. When possible, the number 

 of colonies under each square should actually be counted. 

 As a rule, however, it is customary to count the number of 

 colonies found under each of 6, 8 or 10 squares selected at 

 random from over the surface of the plate. The average 

 number present in one square is then ascertained. 



Fig. 57. Wolffhugel's apparatus for counting colonies. 



The total number of colonies on the plate is found by 

 determining the number of square centimeters which the gel- 

 atin on the plate covers, and multiplying this figure by the 

 average number of colonies per square. Since each colony 



