COOLING HOT-BOTTLED PASTEURIZED MILK. 17 
BY FORCING THE AIR FROM THE TOP DOWNWARD. 
As it is natural to expect, the results obtained by admitting the 
air at the top of the box, forcing it downward through the stacks of 
crates, and out near the bottom of the box, show that it is by far 
the best method. With this arrangement the liquid in the upper 
part of the bottles is cooled first, and of course, owing to its greater 
density, settles to the bottom of the bottles, allowing the warmer and 
therefore lighter liquid to rise to the top and take its place. The 
convection currents are thus taken advantage of, with the result 
that the extreme difference in temperature between the lower quarts 
and the upper pints is only a few degrees. In view of this fact 
the possibility of some of the bottles being insufficiently cooled, 
while others in extreme cases may be frozen, is remote when this 
arrangement is used. Furthermore, the rate of cooling was found 
to be considerably increased by this method as compared with the 
other two methods. In addition to giving a more nearly uniform 
temperature throughout the stack, the variation in the temperature 
of different parts of the same bottle was more nearly the same when 
the air was forced downward than when blown upward. 
From the temperature curves in figure 11 it will be noted that 
the incoming air is at an average temperature of 40° F. (4.4° C), and 
that with an air velocity of 2,512 feet a minute the time required 
to reduce the temperature of the lower quart to 50° F. (10° C.) was 
2 hours and 10 minutes. The average difference in temperature 
between the lower quart and the upper pint was only 8.9° F. (4.9° C). 
In this experiment, as in others, there was some variation in the initial 
temperature of the liquid before cooling was commenced. In this 
case the variation amounted to 8° to 10° F. (4.4° to 5.6° C). 
The curves in figure 12 show more plainly the results of cooling 
by blowing air downward through the stack than those in figure 11, 
for in the former case the initial temperature of all bottles is prac- 
tically the same. The average difference in temperature between 
the lower quart and upper pint (curves 3 and 4, respectively, figure 
12), representing extreme conditions both as to size and location 
of bottles, was only 2.49° F, (1.4° C), while the maximum difference 
in temperature between these bottles was only 4.5° F. (2.5° C). 
There was a difference of only 2° F. (1.1° C.) between the average 
and the maximum, showing that the cooling throughout the stack 
was practically uniform and that with this method all danger that 
some of the bottles are not thoroughly cooled is eliminated. If the 
temperature is taken in any one bottle it is safe to assume that the 
temperatures of the others are within a few degrees of the one taken 
when the crates are arranged as in the experiment. The average 
temperature of the incoming air was about 30° F. (—1.1° C), and 
55377°— Bull. 420—16 3 . 
