COOLING HOT-BOTTLED PASTEURIZED MILK. 33 
the addition of the bunker and heating coils. The crates were stacked 
in the box as shown, and the air was forced by the fan from the 
heating-coil bunker into the top of the box, down through the stack 
of crates, and out at the bottom. The floor of the box was made 
practically air-tight around the crates, so that the hot air necessarily 
had to flow through the crates and around the bottles in its passage 
from, the box. 
In figure 17 it will be noted that the galvanized piping for air was 
so arranged M-.at by manipulating dampers a closed hot-air circuit 
would be maintained for heating, or cold air from the outside could 
be admitted to the box for the purpose of cooling. By closing the 
, dampers in inlet pipe G and outlet pipe B and opening the damper in 
pipe E a closed hot-air circuit would be maintained in which the 
fan A would draw the hot air from the bottom of the box and force 
it upward around the heating coils and into the top of the box 
through the opening H, thence downward through the crates. 
After heating the liquid to the desired point, the steam supply was 
cut off, the dampers were reversed, and cold outside air admitted to 
the box for cooling; that is, the dampers in pipes G and B were 
opened, and the one in pipe E closed. Then cold air from the out- 
side was drawn in by the fan A, forced up through the heating coils 
and into the box through the opening H, thence downward through 
the crates, and outside through the pipe B. The heating coils were so 
arranged that the number of coils could be varied, thus making it 
possible to regulate the amount of heat in the box. The steam pres- 
sure and the amount of condensed steam from the coils were recorded 
during the experiments. 
RESULTS OF HEATING FROM BOTTOM UPWARD. 
Before the equipment was arranged as shown in figure 17, the 
heating coils were placed in the bottom of the box, under the plat- 
form, and the air was forced by the fan around the coils and upward 
through the stack of crates, leaving the box near the top. After the 
liquid was heated to the desired point, cold air for cooling was blown 
downward through the crates. The results obtained under these 
conditions are shown by the curves in figure 18. It will be seen that 
there was a wide variation in the temperatures of bottles at different 
parts of the stack. 
After heating there was an average difference of 16° F. (8.9° C.) 
between the top and bottom quarts of milk, and a maximum differ- 
ence of about 30° F. (16.7° C). This variation in temperature is 
considered too great for satisfactory pasteurization, as certain bottles 
will be overheated while others will not be heated to the required 
point; furthermore, there is a variation in temperature of from 10° 
