53 
Compared with the previous methods this one produces a fairly 
large quantity of formaldehyde gas in the air of the room when used 
during warm weather, say at a temperature of 75° F The results 
under these conditions show that the quantity increases during a rela- 
tively long period of time. In experiment II c, Table 5, the quantity 
of formaldehyde gas found after an interval of 22 hours w T as only 2.75 
per cent less than that found in II a, 90 minutes after closing the 
room. During cold w T eather, say temperature of 43.5° F., very little 
formaldehyde is liberated, as the result of experiment III shows. 
Evaporation takes place very slowly and the formalin polymerizes on 
the sheets. 
Germicidal properties . — In these experiments the organisms were 
prepared and exposed in the usual way, but not until one hour after 
spraying. This allowed time for at least a portion of the formalde- 
hyde to be given off and become diffused. 
A summary of conditions and results is given in each of the following 
experiments : 
EXPERIMENT NO. 28. 
Sheet spraying. 
Formalin, 600 c. c. 
Capacity of room, 2,000 cubic feet. 
Quantity of formaldehyde used per cubic foot, 0.107 gram. 
Quantity of formaldehyde found per cubic foot, 0.02576 gram. 
(Determinations made by drawing air from room beginning 1 hour 26 minutes after spraying 
sheets and consuming 1 hour 15 minutes in the drawing.) 
Yield of formaldehyde, 24.07 per cent. 
Condition of weather: Very light wind, slightly cloudy. 
Tempera- 
ture. 
Relative 
humidity. 
Absolute 
humidity 
(grams per 
cubic foot) . 
Before spraying 
° F. 
76.5 
Per cent. 
72 
7.05 
1 hour after 
76.0 
80 
7.72 
2 hours after ' 
77.0 
81 
8.06 
Organisms not exposed until 1 hour after spraying sheet. 
[4- means growth; — no growth.] 
• 
Organism. 
How exposed. 
Time of exposure in minutes, 
and results. 
5 
10 
20 
30 
60 
90 
120 
B. pyocyaneus 
Filter paper 
+ 
-f 
+ 
B. coli communis 
do 
+ 
+ 
Do 
YTlson method . 
+ 
4- 
+ 
B. dysenteric 
Filter paper 
+ 
+ 
B. subtilis 
do 
+ 
+ 
+ 
Do 
Wilson method 
+ 
+ 
+ 
+ 
+ 
i 
