10 BULLETIN 420, U. S. DEPARTMENT OF AGRICULTURE. 
described. Several experiments were carried out, using air at different 
temperatures for cooling purposes. 
In figure 6 are shown the results of cooling by air 'at an average 
temperature of 67.9° F. (19.9° C). A complete set of curves is 
incorporated, principally to show the rate of cooling with air at this 
high initial temperature. Curves 3 and 4 show the relative rate of 
cooling between quart bottles of milk and water, respectively, when 
placed side by side at the bottom of the stack. These curves agree 
very closely with those in figure 2. Curves 5 and 6 show, respectively, 
the rate of cooling between pints and quarts placed at the top, or 
fifth, tier of crates. Curves 3 and 6 are plotted from temperature 
readings taken in the lower quart of milk and the upper quart of 
water, respectively. The average difference hi temperature between 
these bottles was 9.8° F. (5.4° C), while the maximum difference was 
about 19° F. (10.5° C). Curves 5 and 6 are practically parallel 
throughout the range, the curve showing the drop rn temperature of 
the pint remaining throughout about 5 degrees lower than that of 
the quart. The average difference in temperature between curves 4 
and 6, which are plotted from temperature readings taken in the 
lower and upper quarts of water, respectively, is 10.5° F. (5.8° C), 
while the greatest is 20° F. (11.1° C). 
The temperatures indicated by the curves in figure 7 were deter- 
mined in a pint and a quart bottle at the top of the stack and also one 
of each kind at the bottom. The temperature of the lower quart and 
pint, as shown by curves 3 and 4, respectively, drops below that of 
the outgoing air. It should be remembered, however, that the tem- 
perature of the outgoing air was taken at the point where it left the 
box; hence, the temperature of the air passing over the bottles was 
lower than that indicated by the curves. Toward the end of the run 
the temperature of the lower pint was reduced below that of the 
incoming air. This was caused by the temperature of the outside air 
rising so rapidly that that of the lower pint, which had been cooled, 
had not had sufficient time to adjust itself to the rising temperature 
of the outside air. The average difference between the lower pint 
and the upper quart, curves 4 and 5, was about 25° F. (13.9° C), 
while the maximum difference in temperature was about 43° F. 
(23.9° C). This difference becomes more marked when the tempera- 
ture of the outside air is lower, consequently, when blowing air up 
through the crates there is great danger of the top crates not being 
cooled properly, while on the other hand the milk in the lower crates 
may be frozen, provided the initial temperature of the incoming air 
is below the freezing point. By comparing temperature curves 3 and 
6, lower quart and upper pint, respectively, it will be noted that the 
temperatures in the two bottles were much nearer each other. The 
maximum difference was about 12° F. (6.7° C), but the average was 
