250 
to be examined, thus making a dilution of 1 to 100. From it 1 cubic 
centimeter was removed and added to the second flask containing 9 
cubic centimeters, making a second dilution of 1 to 1.000. By a con- 
tinuance of this method, namely, the removal of 1 cubic centimeter 
and its addition to the fresh flask containing 9 cubic centimeters of 
pure water, the dilutions may be rim as high as desired. For the 
routine of this work dilutions of 1 to 1.000 : 1 to 10.000 ; 1 to 100.000, 
and 1 to 1.000.000. were adopted. 
The sowing of the organisms on the nutrient jelly was made by the 
removal of 1 cubic centimeter from the flask containing the desired 
dilution and its transference to a sterile petri plate, into which was 
immediately poured the melted medium and the organisms evenly 
distributed by shaking with a rorary motion. Duplicate plates were 
made in all cases and 2 per cent lactose agar was selected as the 
nutrient medium affording most satisfactory results. All the plates 
for this investigation were grown at a temperature of 30 = C. for a 
period of three days, after which the colonies when numerous were 
counted by means of a Stewart counting chamber or when but few 
by the naked eye alone. 
The presence of gas-producing organisms was determined in this 
investigation by adding 1 cubic centimeter of the 1 to 100 dilution to 
sterilized 2 per cent dextrose fermentation tubes and incubating at 
30° C. for three days. IVhen gas formation took place the quantity 
was estimated by the ruled scale method, as described by Frost in his 
Laboratory Manual. 
An endeavor was made to determine systematically the presence 
and approximate number of streptococci in each sample of ice cream, 
cream and milk , which has been examined recently by this Depart- 
ment. For this purpose 15 cubic centimeters were centrifugalized 
with an electric centrifuge for a period of fifteen to twenty minutes, 
and from the sediment were made several smears which were stained 
with methylene blue. Such a procedure yielded results with milk and 
cream alone, but when in the form of ice cream, especially those with 
fruit or chocolate flavors, the debris seemed to interfere to such an 
extent that satisfactory results were not always obtained. TTith the 
vanilla flavors the results were better, but even in such cases they were 
exceedingly rough. Hence there are a number of blanks in the tables 
and summaries of this work dealing with the presence in ice cream of 
streptococci, and the determination of the number of leucocytes per 
cubic centimeter in ice cream was made in but a few cases. 
Between October 13. 1906. and July 29. 1907. 263 samples of ice 
cream, collected in the City of TC ashington. were investigated as above 
outlined. That the bacterial flora in the majority of these ice creams 
