ch a point that when the motor is
stopped the amount of oxygen to be added to bring the tension to a
definite point will be small.

Under these conditions it is hoped to secure a more satisfactory
comparison of the analyses as made by means of the Sonden apparatus and
as calculated from the composition of the residual air by the
gravimetric analysis. It remains a fact, however, that no matter with
what skill and care the gasometric analysis is made, either
gravimetrically or volumetrically, the calculation of the residual
amount of oxygen presents the same difficulties in both cases.

CALCULATION OF TOTAL OUTPUT OF CARBON DIOXIDE AND WATER-VAPOR AND OXYGEN
ABSORPTION.

From the weights of the sulphuric-acid and potash-lime vessels, the
amounts of water-vapor and carbon dioxide absorbed out of the
air-current are readily obtained. The loss in weight of the oxygen
cylinder increased by 0.4 per cent (see page 88) gives the weight of
oxygen admitted to the chamber. It remains, therefore, to make proper
allowance for the variations in composition of the air inside the
chamber at the beginning and end of the different periods. From the
residual sheets the amounts of water-vapor, carbonic acid, and oxygen
present in the system at the beginning and end of each period are
definitely known. If there is an increase, for example, in the amount of
carbon dioxide in the chamber at the end of a period, this increase must
be added to the amount absorbed out of the air-current in order to
obtain the true value for the amount produced during the experimental
period.

A similar calculation holds true with regard to the water-vapor and
oxygen. For convenience in calculating, the amounts of water-vapor and
carbon dioxide residual in the chamber are usually expressed in grams,
while the oxygen is expressed in liters. Hence, before making the
additions or subtractions from the amount of oxygen admitted, the
variations in the amount of oxygen residual in the system should be
converted from liters to grams. This is done by dividing by 0.7.

CONTROL EXPERIMENTS WITH BURNING ALCOHOL.

After having brought to as high a degree of perfection as possible the
apparatus for determining carbon dioxide, water, and oxygen, it becomes
necessary to submit the apparatus to a severe test and thus demonstrate
its ability to give satisfactory results under conditions that can be
accurately controlled. The liberation of a definite amount of carbon
di