TECHNIC OF BACTERIOLYTIC TESTS 



353 



All plates are incubated at 37 C., for from twenty-four to thirty-six 

 hours and the colonies then counted. In some plates the number may 

 be so large that counting will be inaccurate and unnecessary. Sig- 

 nificance can be attached only to marked and easily recognizable dif- 

 ferences. According to Neisser, the growth is best and most rapidly 

 described by means of approximate estimates, using a scheme somewhat 

 like the following: or almost none; about 100; several hundreds; 

 thousands; very many thousands; infinite numbers. A distinct bac- 

 tericidal action is then present only if the controls react normally, and 

 if a reduction of colonies from an infinite number or many thousands to 

 or very few has occurred. As previously stated, the test can then be re- 

 garded as satisfactory only if the lower limits of bactericidal activity have 

 been reached and the last plates show an increase in the number of colonies. 

 Examination of these plates is very much facilitated by using a colony 

 counter, that of Stewart being particularly serviceable with agar plates. 



The controls are first examined and the results recorded, and finally 

 those of the patient's serum are set down. 



As a practical illustration, the results of a test with a rabbit typhoid 

 immune serum are given, because this is in general an average example 

 of those usually obtained: 



23 



