12 BULLETIN 1022, U. S. DEPARTMENT OF AGRICULTURE. 
These curves show how, as the sealing temperature is increased, 
the pressure is developed and bring out the fact that as the cooling 
temperature approaches 100° C. the vacuum increases progressively. 
They also show how the vacuum is affected by the temperature to 
which it is cooled. For example, a can sealed at 70° C. and subse- 
quently cooled to 20° C. should give a theoretical vacuum of about 
12.3 inches; if cooled to 30° C., a vacuum of about 10.9 inches; or if 
cooled to 0° C., a vacuum of 14.3 inches. They likewise illustrate 
how cans of food having a low vacuum, although sound and health- 
ful, may, when stored in warm places or shipped to hot climates, 
show no vacuum at all, or possibly in some cases even develop spring- 
ers. For instance, a can of food which at 10° C. showed approxi- 
mately 3 inches of vacuum when held at 30° C. would have no vacuum 
at all. Similarly, though of less importance, it shows how cans 
having relatively high vacuums would suffer more in shipment and 
handling in low temperatures, due to the denting or collapsing of 
the cans. One may, therefore, gain from these curves an idea of the 
approximate vacuums to be expected under- conditions within the 
range of canning operations. 
EXPERIMENTS WITH DISTILLED ‘WATER FOR COMPARISON. 
As is well known in canning, theoretical values are practically 
never obtained because numerous variable factors enter in which in- 
fluence practical results. In order to discover what and how great 
these variable factors are, to learn how closely the theoretical and the 
practical results agree, and to obtain information which would serve 
as a base line for the comparison of various food materials when 
canned in different ways, preliminary pressure and vacuum experi- 
ments were made with water, using No. 2 and No. 3 tin cans. The 
experiments were conducted as ieee The test cans were prepared 
by carefully soldering all seams and joints to insure against leakage: 
the required quantity of water was measured in; the can was placed 
in the retort in the manner illustrated and held under carefully con- 
trolled conditions until both can and contents came to the desired 
uniform temperature throughout; the can was then sealed. The 
barometric reading at the time of sealing was noted and the pressure 
determinations immediately performed. In this way tests were made 
using various quantities of water, sealed at different temperatures, 
and processed in the retort at 100°, 109°, 116°, and 121° C. 
PRESSURE STUDIES. 
Pressure readings obtained in tests—Table 1 gives the resulis of 
maximum-pressure tests upon No. 3 cans with 100 c. c. and with 950 
¢. c. of water. 
