CALCULATION OF RESULTS. 2/9 



First the total number of bacteria upon any plate is 

 counted. Then the plate is studied and the total number of 

 each recognized type is determined by actual count. The sum 

 of these numbers of each species should give a number equal 

 to that of the total number on the plate, provided all of the 

 colonies on the plates are easily differentiated. It usually 

 happens, however, that the total number of differentiated 

 colonies fails to add up as much as the total number of bac- 

 teria on the plate. This, of course, means that some of the 

 colonies are not clearly recognized types and the difference 

 between the total number and the sum of the numbers of the 

 recognized species must be regarded as undetermined or mis- 

 cellaneous. The number of miscellaneous colonies is found 

 by experience to vary greatly, according to circumstances. 

 If the liquefiers are abundant, and the plates must be studied 

 before they are four days old, the number of miscellaneous 

 colonies that will be found is commonly large, whereas if the 

 plates can be kept till six days old the number of the undeter- 

 mined colonies is small. Other circumstances interfere occa- 

 sionally to prevent the satisfactory differentiation, and when- 

 ever this occurs the percentage found in the undetermined 

 column always rises. The number found in this column, 

 therefore, is usually an indication of the success of the proper 

 differentiation. 



The results which are obtained should then be tabulated 

 and each table should be a double one. It should contain first, 

 the total number of each type of bacteria detected, and 

 second, the percentage of each type. The purpose of the 

 latter is to show the relative preponderance of the different 

 microorganisms. The advantage of this is considerable. 

 From facts already mentioned it follows that if the milk 

 shows large numbers of bacteria, but of this large number the 

 great per cent, are lactic organisms, the milk must be re- 

 garded as normal though rather old. If, on the other hand, 

 the results should show a large per cent, of liquefiers, or even 



