6 BULLETIN 420, U. S. DEPARTMENT OF AGRICULTURE. 
The bottles placed in still air were cooled by natural circulation 
brought about by the difference in temperature between the compara- 
tively warm air in contact with the outer surface of the bottles and 
that at a distance. As the air in contact with" a hot surface, such as 
that of the bottles containing milk, becomes heated, it rises and its 
place is taken by cold air; hence, the rate of circulation of the air 
depends upon the difference in weight of the respective columns of 
warm and cold air. As this difference in weight is very slight, the 
circulation is naturally slow, and hence a slow rate of cooling follows. 
iso t 
— ■ 
\ 
*M 
^ 
\ 
. 
fc 
s 
tJ Bt> 
^ 
s 
8? 
I 70 
s 
^ 
^ 
^ 
-^ 
r 
r' 
a;' 
=>. 
ae 
ze t 
i 
i 
3 
a 
o 
-3. 
O 
<?- 
o 
s 
o 
6 
j 
7 
T 
O 
a 
- A 
A 
o 
.' 
M 
!' 
a 
a 
10 
■' 
30 
A 
■M 
A 
UJ 
Fig. 2. — Cooling of milk and water by forced air.— Difference in the rate of cooling of milk and water when 
exposed to an air blast. Quart bottles. 
In the case of forced circulation of air, however, the film or sleeve 
of hot air surrounding the vessel is broken up, and a large volume of 
cold air is brought into direct contact with the outer surface of the 
container; the cooling therefore goes on very much faster. The 
forced circulation of the air hi these experiments was made by a 16- 
inch electric fan, which produced over the bottles an air velocity of 
approximately 1,250 feet a minute, measured by an anemometer 
which had just been calibrated by the Bureau of Standards. 
It will be noted by reference to figure 3 that it required 4J hours 
to reduce the temperature of the milk in the pint bottles when placed 
in still air from 142° to 50° F. (61.1° to 10° C.) with a practically con- 
stant temperature of 40° F. (4.4° C.) inside the room. When the air 
