184 STATE BOARD OF AGRICULTURE 



A Study of Methods jor Bacterial Analyses of Market Milk. 



Bacterial counts of milk made upon different kinds of media vary 

 consideralbly in number present per cviibic centimenter. This is due, 

 largely, to the composition of the media used. As one kind of medium 

 is most suitaible for the growth of certain types of bacteria it will give 

 higher counts than other media Avhen that particular type of organism 

 predominates. If another type or organism predominates in the milk 

 another medium may be exx)ected to give the highest counts. 



It occasionally happens that a sample of milk with a count of 100,000 

 bacteria per cubic centimeter upon one medium will have a count of 

 only half that num(b6r upon another medium. With other samples 

 of milk the reverse may happen. Usually, however, certain media tend 

 to give the highest counts, while other media tend to give the lowest 

 counts. 



The usual reason for making bacteriological counts upon milk is to 

 get an idea of how carefully the milk has been handled up to the time 

 that the sample was taken. If the l)acteriological count is low we as- 

 sume that the milk has been handled in such a careful way that good 

 keeping quality and reasonalble safety are likely. If the count is very 

 high we suspect that the milk has been so carelessly handled that good 

 keeping quality cannot Ibe expected and safety from disease producing 

 organisms is in dou'bt. 



As a member of the committee upon methods of bacterial analyses of 

 milk and milk products appointed by the International Association of 

 Dairy and Milk Inspectors, I undertook in 1921 to show the value of 

 plate counts o^btained upon four kinds of agar commonly used in milk 

 work and to show the relation of the counts to the pH score. (1) 



Early results of this work have been included in the committee re- 

 ports (2) 1»ut my Avork has been continued and extended. 



Tlie methods used in this work are tlie 1921 Standard Methods for 

 Bacteriological Examination of Milk, published by the American Public 

 Health Association. Bacto Agar used in Group No. I and II is out of 

 Lot No. 10415 which was furnished by tlie Digestive Ferments Company 

 to mem'bers of the committee. Other Bacto Agar w^as out of regular 

 stock. Milk-powder agar was the milk-powder Agar A described by 

 Ayers and Mudge. (3.) 



Media were adjusted to pH G.5-G.8 Avhere necessary. It was not neces- 

 sary to adjust the reaction of Bacto Agar and Agar A. 



Group No. I includes 36 samples of milk, 18 of which were taken upon 

 a warm day and tested at once. The other 18 samples were taken from 

 the same patrons' cans a few days later, but upon a very cool day. 



Group No. II includes 48 samples of milk, 21 of w^hich were plated 

 upon four kinds of media and pH score determined. The samples were 

 then incubated at i-oom temperature for two hours and reteste<l. 



Grou]) No. Ill includes samples of milk plated upon different media 

 and incubated at 37° 0. for one hour and retested. 



I assume that the samples of milk in Groups II and III when platerl 

 the second time slionld be in poorer condition bacteriologically tJian 

 when they were tested the lirst time. If this is true the method which 

 shows the samples in poorer condition most consistently when tested the 

 secoiicl time would be the best method. 



