438 
After careful consideration of the subject the following methods 
have given satisfactory results in this laboratory: a 
The samples were always collected in the original containers, 
either pint or quart bottles being purchased for our purposes. Some 
of these samples were obtained from the wagon on the street, others 
from the dairy, and still others were obtained from houses in various 
parts of the city, at once after being delivered in the usual course of 
trade. It is therefore believed that the samples examined fairly rep- 
resent the average milk obtained by the householder. The samples 
were collected early in the morning and at once placed in a metal con- 
tainer filled with cracked ice. From six to eight samples were collected 
each morning from various parts of the city, and rarely more than two 
hours elapsed from the collection of the first sample to the time it was 
received in the laboratory. The temperature was taken with a good 
thermometer at the time the sample was collected, but always from 
a different bottle, which was afterwards used for chemical purposes. 
It was noted that after the milk stood on ice for some time that 
there might be a difference of 6 to 8 degrees between the top and the 
bottom layers of the milk in a pint bottle. The milk was always 
shaken well in order to mix the cream and to help break up the bac- 
terial clumps before the bottle was opened, which was done with the 
usual bacteriologic precautions. For ordinary market milk the fol- 
lowing dilutions were made : 
1 cubic centimeter milk + 99 cubic centimeters sterile water. 
0.1 cubic centimeter of this was used for the first plate, which rep- 
resented 1 : 1,000. 
0.5 cubic centimeter of the first dilution was then added to 49.5 
cubic centimeters of sterile water. One cubic centimeter, of this dilu- 
tion when plated represented 1:10,000, and 0.1 cubic centimeter of 
this dilution represented 1 : 100,000. 
The dilutions were vigorously shaken at least twenty-five times in 
accordance with the standard methods for water analysis in order to 
obtain uniform suspension of the bacteria. Sterile distilled water was 
used as a diluent. 
The final dilution was measured directly into a petri dish and agar 
poured at a temperature of between 40° and 45° C. in the usual way- 
After the plates were well set, they were grown at 37° C., which 
temperature appears not only to favor the maximum growth of 
bacteria ordinarily found in the milk, but has the additional advan- 
tage of favoring the kinds of bacteria belonging to the pathogenic 
a Since writing this article the committee on standard methods of bacterial milk 
analysis have presented a preliminary report, which appeared in the American Journal 
of Public Hygiene, vol. 17, November, 1907, pp. 331-364. At the Winnipeg meeting 
of the American Public Health Association in September, 1908, the committee pre- 
sented a report of further progress, an abstract of which appeared in the American 
Journal of Public Hygiene for November, 1908, p. 425. (See also Heineman and 
Glenn’s recent article on “A comparison of practical methods for determining the 
bacteriological content of milk,” Joum. Infectious Diseases, vol. 5, Oct. 20, 1908, pp. 
412-420.) 
