EXAMINATION OF AIR, WATER, AND SOIL. 131 



germs which do not grow on ordinary media. Large numbers 

 of colonies of bacteria soon make their appearance. 



Besides determining the actual presence of these germs, 

 quantitative tests must be made. Various methods have been 

 suggested : 



Hesse draws a current of air through a glass tube 70 cen- 

 timeters long and 3.5 centimeters wide, which is coated on 

 the inside with a film of gelatin, like an Esmarch roll culture. 



Petri uses small sand filters placed in a wide glass tube 

 (Fig. 38). The sand is first sifted, then sterilized by heat, 



FIG. 38. 



Petri's apparatus for bacteriologic analysis of air: a, tube packed with sand; 

 arrows indicate entrance and exit of air current. 



after which it is placed in the tube, supported by small wire 

 baskets. Two such filters are placed in one tube. One end 

 of the tube is closed with a cork, through the centre of which 

 passes a thin glass tube. The entire apparatus is then steril- 

 ized in the hot-air sterilizer. By means of an air-pump, 

 100 liters of air are aspirated through the sand in from ten 

 to twenty minutes. The sand filters are then placed in sterile 

 dishes containing gelatin, and the colonies are counted as 

 they develop. The objection to this method is that the sand 

 granules are apt to be mistaken for colonies. 



Tucker and Sedgwick (Fig. 39) have improved this method 



