286 Trans. Acad. Sci. of St. Louis. 
bation, the reaction of the medium being + 1.5%, and 12% 
gelatin being used in each case. For some reason in this 
series, the samples from Lockport, Joliet and Peoria showed 
few liquefying bacteria on the third day of incubation, while 
those from Alton and lower down along the stream showed 
not only more liquefiers, but the liquefaction of the gelatin 
had usually progressed to a considerable extent. At a cer- 
tain period of the work I was stationed at Joliet, Ill., and 
samples from that district could be plated soon after collec- 
tion. In these cases, I found a large percentage of liquefying 
colonies so that the low averages as shown in the table may 
be explained by the condition of transportation of the sample. 
The quantitative estimation as shown by plate cultures 
should always be considered as relative and not as giving the 
absolute number of bacteria, for it has been found that 
various media used for plating give widely different results. 
The nature of the nutrient media, the conditions of growth, 
the presence or absence of air, moisture, etc., are all im- 
portant factors modifying the actual bacterial number. To 
demonstrate what effect the absence of air would have on the 
number of bacteria, plates were prepared in the regular way 
and similar sets of plates were grown under anaerobic condi- 
tions with the following results : — 
_) SABLE &. 
Bacteria per c.c. 
Conditions of Growth. P 
Source of Sample. Date. thal 
Aerobic. | Anaerobic 
Des Plaines R., Lockport.....| Oct. 21, 1901. 1,200 650 54% 
Bear Trap Dam, eT Cor hd 93,000 50,100 54% 
Des Plaines R., “ vine o Ock, 81, 1901. 1,900 900 49% 
Bear Trap Dam, by obs c 332,000 175,000 53% 
With the removal of oxygen but half the number of 
bacteria developed, and it is important to consider this fact in 
connection with sewage polluted water. ‘The quantitative 
examinations, while not giving absolute results, are often of 
value when the numbers run very high as an indication 
