PHYSIOLOGICAL AND BIOCHEMICAL TECHNICS 171 



be plated at least in triplicate. After the colonies have been counted, the 

 number of cells in the original suspension is calculated. Alternately, the 

 inocula may be spread over the surface of agar plates in the case of 

 obligate aerobes. 



In the latter procedure, plates are first poured and allowed to solidify. 

 To remove excess water from the surface of the agar, the plates are dried 

 overnight at 37°C or for 2 hr at 50-55°C. The sample to be counted is 

 deposited (0.1 ml in 3-4 separate drops) on the agar surface and spread 

 uniformly by means of a sterile bent glass rod. Since excess surface 

 water has been removed, difficulties due to colony spreading are obviated. 



A culture in the maximum stationary phase of the growth curve will 

 often contain about 10^ cells per ml; thus, the suitable dilution for plating 

 will be found in the seventh serial dilution tube for a fully grown culture. 

 The sixth and eighth tubes should also be plated in order to allow for 

 errors in estimation of the original culture population. 



It should be obvious that not all bacterial species mil grow on nutrient 

 agar and will therefore require a different agar medium in order to 

 develop colonies. Anaerobic species must, of course, be incubated under 

 anaerobic conditions. Furthermore, some species will require different 

 diluent fluids, since saline may allow lysis of some cells, thereby resulting 

 in low counts. 



The method is based upon the assumptions that each cell develops into 

 one colony and that each colony is derived from only one cell. These 

 assumptions are not necessarily valid for those species which grow in 

 chains or clumps. Since not all cells are capable of reproduction, each 

 cell mil not, as assumed, yield a colony. For these reasons, the estimate 

 of number of cells will nearly always be lower than the true value. In 

 addition, the proportion of nonreproducing cells may vary at different 

 stages of growth; thus the error will not be constant. Inconstancy of 

 error also will obtain when the size of the clump or the length of the chain 

 varies with age of the culture. 



For a discussion of the application of statistical method to bacterial 

 enumeration procedures, see the review by Stearman (1955) and the pre- 

 vious reports cited therein. 



Direct microscopic method. A counting chamber of the hemocytometer 

 type is employed in this method. The chamber consists of a ruled slide 

 and a cover slip constructed in a manner such that a definite known 

 volume is delimited by the cover slip, slide, and ruled lines. Detailed 

 description and instructions for use are furnished with the commercially 

 available apparatus. The Petroff-Hauser or Helber bacterial counting 

 chamber gives results superior to those obtained with the ordinary hemo- 

 cytometer. The Petroff-Hauser slide is ruled in 0.05-mm squares, with 

 the distance between the cover slip and slide equal to 0.02 mm. 



