SOME OBSERVATIONS ON THE EFFICIENCY OF THE 

 PRESENT STANDARD AGAR FOR THE ESTI- 

 MATION OF BACTERIA IN MILK 



H. J. SEARS AND LULU L. CASE 



Department of Public Health, Berkeley, California 

 Received for publication November 27, 1917 



On receipt of the last report of the Committee on Standard 

 Methods^ the workers in this laboratory noted with surprise 

 and also with some skepticism that the content of nutrient sub- 

 stances in the standard agar had been reduced still further. We 

 accepted the revision of methods, however, in their entirety and 

 have been adhering to the new standards ever since but with 

 an attitude somewhat more critical than heretofore. Whether 

 it was due to this critical attitude or whether it was on account 

 of the change in the composition of the standard agar, it is a 

 fact that we soon noticed inaccuracies and inconsistencies in some 

 of our results which were too important to be overlooked. 



The chief difficulty which we have encountered, and for which 

 we are constrained to blame the present standard agar, consists 

 in the failure of the counts made from the different dilutions of 

 the same sample to check with each other. It was not infre- 

 quently noticed, for example, that the plates from the 1-100 dilu- 

 tion would give the sample a total count many times greater 

 than those from the 1-1000 dilution. Very often it would be 

 quite impossible to count the former while the latter would show 

 fewer than the minimum of 20 colonies. This sort of a situation 

 could only leave the worker the choice of estimating the number 

 of colonies on the very much overcrowded 1-100 plates or of 

 reporting the number indicated by the obviously unreliable 1- 

 1000 plates. 



^ Provisional Report of the Committee on Standard Methods of Bacteriologi- 

 cal Analysis of Milk, 1916, Am. Jour. Pub. Health, 6, 1315. 



531 



THE JOURNAL Or BACTEEIOLOaT, VOL. Ill, NO. 6 



