THE DETERMINATION OF BACTERIA IN ICE CREAM. 
15 
Table VIII. — Colony counts on duplicate plates from the same dilution of a sample of 
ice cream. 
Sample 
No. 
Number of colonies on duplicate plates. 
Varia- 
tion. 
1 
2 
3 
4 
5 
6 
7 
8 
1 
2 
3 
4 
5 
243 
450 
321 
319 
144 
229 
460 
302 
310 
152 
225 
510 
307 
267 
153 
208 
470 
300 
298 
178 
232 
480 
310 
280 
178 
218 
470 
Percent. 
16.8 
26.6 
7.0 
19.4 
23.6 
500 
570 
INTERPRETING DIFFERENCES IN BACTERIAL COUNTS. 
When expressing bacterial counts the mere statement of the dif- 
ference in the number of bacteria between two samples is not suffi- 
cient; in fact, it is impossible to interpret such differences properly 
without taking into consideration the total number of bacteria 
found in each sample. If, for example, it is stated that 1 cubic 
centimeter of ice cream contains 75,000,000 more bacteria than 
another sample, the difference of so many millions conveys to the 
mind a marked disparity in the two samples. If, on the other hand, 
there is said to be a difference of only 750 or 7,500 bacteria per cubic 
centimeter, we should immediately think of the samples as being 
practically the same. 
Table IX. — Hypothetical statement showing variable differences in bacterial counts of 
two samples of ice cream having a fixed ratio between colony counts and a fixed per cent 
, of variation. 
Colony 
375 
300 
375 
300 
375 
300 
375 
300 
375 
300 
375 
300 
375 
300 
Dilution. 
JltolO. 
ktoioo... 
}l to 1,000 
il to 10,000 
1 to 100,000.... 
1 to 1,000,000.. 
}l to 10,000,000. 
Bacteria 
per c. c. 
3,750 
3,000 
37,500 
30,000 
375,000 
300,000 
3,750,000 
3,000,000 
37,500,000 
30,000,000 
375,000,000 
300,000,000 
1,750,000,000 
L3, 000, 000, 000 
Difference 
in count. 
750 
7,500 
75,000 
750,000 
7,500,000 
75,000,000 
■750,000,000 
Variation 
between 
counts. 
Per cent. 
25.00 
25.00 
25.00 
25.00 
25.00 
25.00 
25.00 
Let us consider a hypothetical case, as shown in Table IX. We 
will assume that one sample of ice cream shows an average colony 
count of 375 and another 300. With a dilution of from 1 to 10, 
there would be a total count of 3,750 in one sample and 3,000 in 
the other, a difference of 750 bacteria per cubic centimeter, and a 
variation of 25 per cent between counts. If the same colony count 
were from a dilution of from 1 to 100 the difference in the bacteria 
in the samples would be 7,500, \ the variation would be still the 
