required.

VARIATIONS IN THE APPARENT VOLUME OF AIR.

In the earlier form of apparatus the largest variation in the apparent
volume of air was due to the fluctuations in the height of the large
rubber diaphragms used on the tension equalizer. In the present form of
apparatus there is but one rubber diaphragm, and this is small,
containing not more than 3 to 4 liters as compared to about 30 liters in
the earlier double rubber diaphragms. As now arranged, all fluctuations
due to the varying positions of the tension-equalizer are eliminated as
each experimental period is ended with the diaphragm in exactly the same
position, _i. e._, filled to a definite tension.

In its passage through the purifiers the air is subjected to more or
less pressure, and it is obvious that if these absorbers were coupled to
the ventilating system under atmospheric pressure, and then air caused
to pass through them, there would be compression in a portion of the
purifier system. Thus there would be a contraction in the volume, and
air thus compressed would subsequently be released into the open air
when the absorbers were uncoupled. The method of testing the system
outlined on page 100 equalizes this error, however, in that the system
is tested under the same pressure used during an actual experiment, and
hence between the surface of the sulphuric acid in the first porcelain
vessel and the sulphuric acid in the second porcelain vessel there is a
confined volume of air which at the beginning of an experimental period
is under identically the same pressure as it is at the end. There is,
then, no correction necessary for the rejection of air with the changes
in the absorber system.

CHANGES IN VOLUME DUE TO THE ABSORPTION OF WATER AND CARBON DIOXIDE.

As the water-vapor is absorbed by the sulphuric acid, there is a slight
increase in volume of the acid. This naturally results in the diminution
of the apparent volume of air and likewise again affects the amount of
oxygen admitted to produce constant apparent volume at the end of each
experimental period. The amount of increase which thus takes place for
each experimental period is very small. It has been found that an
increase in weight of 25 grams of water-vapor results in an increase in
volume of the acid of some 15 cubic centimeters. Formerly this
correction was made, but it is now deemed unnecessary and unwise to
introduce a refinement that is hardly justified in other parts of the
ap