6 BULLETIN 744, U. S. DEPARTMENT OF AGRICULTURE. 
dairy farms have some sort of tank in which water or water and ice are 
used to cool and store milk. According to a recent questionnaire ap- 
proximately 80 per cent of the dairy farms producing market milk 
use some sort of cooling tank. The total number of dairies that re- 
ported approximated 40,000, located in 32 States. Nineteen per cent 
of the tanks were reported to be of metal, 25 per cent of wood, 31 
per cent of concrete, and the construction of the remaining 25: per 
cent was not mentioned. Very few tanks were insulated, and in 
most cases no effort was made to minimize the loss of cooling effect _ 
due to the surface of the tank. When an abundant supply of cold 
running water continually passes through the tank it is unnecessary 
to go to the expense of insulation, but under other conditions, es- 
pecially during hot weather, the loss of cooling effect from an unin- 
sulated tank is considerable. 
The location of the tank with reference to the wind and sun has 
an important bearing on its ability to maintain low temperatures. 
In order to determine the best construction and location for the 
cooling tank, tests were conducted to ascertain the relative milk- 
cooling efficiency of tanks made of galvanized iron, concrete, wood, 
and cork-insulated material, both indoors and outdoors, and with 
and without covers. All the tanks used had the same inside dimen- 
sions, namely, 30 inches in width, 60 inches in length, and 25 inches 
in depth, as shown in figures 3, 4, 5, and 6. The covers were made 
of two layers of $-inch tongue-and-groove boards. 
All tests were made for a period of 9 hours, beginning at 8 a. m. 
and ending at 5 p.m. The average air temperature throughout the 
series of tests was-about 80° F. All the thermometers used were 
compared with a standard thermometer and variations from true 
readings corrected. The thermometer used for taking the air tem- 
perature was protected from the direct rays of the sun. Air and 
water temperatures were taken every hour and before taking the 
ninth-hour reading of the water in the tank it was thoroughly stirred. 
Each tank contained 1,100 pounds of water at an initial temperature 
Or o4°) By 
The conditions of the tests were in brief as follows: In test No. 1 
the tanks were left uncovered and set side by side in the sun. The 
average air temperature for the 9 hours was 84.2° F. In test No. 2 
the tanks were covered and the air temperature averaged 83.5° F. 
In test No. 3 the tanks were sheltered from the sun by a building 
and were uncovered. The average air temperature was 85.2° F. In 
test No. 4 the tanks were sheltered by a building, were covered, and 
at an average air temperature of 83.4° F. Except as noted all con- 
ditions were similar for each test. 
Figures 7 to 10, inclusive, show the curves of the change in tem- 
perature of the air and water under the various conditions of the 
ae ae 
j ras Le iat) POO anc aa aes ain has ene 
tee 
