142 Report oF THE BACTERIOLOGIST OF THE 
process was repeated and a small fraction of a e. c. from the second 
dilution was added to the culture medium. 
Cultures were made in Petri dishes having an internal diameter 
of 91-92 mm. For the sake of convenience in counting and to 
prevent the inhibiting effect of closely-crowded colonies, the aim 
was to so arrange the dilution that the growth would be about 
500 colonies to the plate. In the case of the pasteurized milk, 
no dilution was necessary, but a measured fraction of a cubic 
centimeter was added directly to the nutrient media. 
+ Media used.— The tabulated results given below were all 
obtained upon lactose agar made neutral to phenolphthalein with 
sodium hydroxide and containing 2 per ct. lactose and 1.7 per 
ct. agar. Agar was chosen in preference to gelatin, because in 
some previous work of a similar nature at the Wisconsin Station 
it was found that agar at 28° C. (81.5° F.) gave higher numerical 
results than gelatin at room temperature. Among the substances 
now available, there seems to be none that will call out all the 
individual germs when left at any one temperature. 
It is not maintained by the authors that the numbers given 
below represent the exact number of organisms present either in 
the pasteurized or unpasteurized milk. All that is hoped for is 
that they are a close approximation and that having been taken 
under similar conditions may be found to be directly comparable. 
Incubating temperature.— The plates were placed in an incuba- 
tor at 30° C. (83° F.) and counted at the end of 48 hours. This 
temperature was believed to be near the optimum for the growth 
of most of the germs present and the time was thought to give 
maximum returns with a minimum amount of error. An expos- 
ure at higher temperature caused a rapid drying of the plates and 
one for a longer time did not usually give higher results, while 
the rapid spreading of superficial colonies made the counting un- 
certain. 
Growth at room temperature required so much longer time as 
to complicate the work and the rapid multiplication of certain 
proteus forms made an accurate count very difficult, 
