essary to take special
precautions against the subsidence of the paste. Recourse is therefore had
to another method. In the first place, an aperture is formed in the lower
part of the mould through which the liquid may flow at the desired moment.
Afterward, in order to prevent the solidified but still slightly soft
paste from settling under its own weight at this moment, it is supported
by directing a current of compressed air into the mould, or, through
atmospheric pressure, by forming a vacuum in the metallic jacket in which
the mould is inclosed.

The history and description of these processes have been several times
given, and I shall therefore not dwell upon them, but shall at once
proceed to make known the new points that Mr. Renard has communicated to
me.

The first point to which it is well to direct the manufacturer's attention
is the preparation of the plaster moulds. When it concerns an object of
large dimensions, of a vase a yard in height, for example, the moulder is
obliged to cut the form or core horizontally into three parts, each of
which is moulded separately. To this effect, it is placed upon a core
frame and surrounded with a cylinder of sheet zinc. The workman pours the
plaster into the space between the latter and the core, and, while doing
so, must stir the mass very rapidly with a stick, so that at the moment
the plaster sets, it shall be as homogeneous as possible. In spite of such
precautions, it is impossible to prevent the densest parts of the plaster
from depositing first, through the action of gravity. These will naturally
precipitate upon the table or upon the slanting sides of the core, and the
mould will therefore present great inequalities as regards porosity. Since
this defect exists in each of the pieces that have been prepared in
succession, it will be seen that when they come to be superposed for the
moulding of the piece, the mould as a whole will be formed of zones of
different porosities, which will absorb water from the paste unequally.
Farther along we shall see the inconveniences that result from this, and
the manner of avoiding them.

[Illustration: FIG. 1]

The mould, when finished, is dried in a stove. Under such circumstances it
often happens that there forms upon the surface of the plaster a hard
crust which, although it is of no importance as regards the outside of the
mould, is prejudicial to the interior because it considerably diminishes
its absorbing power. This