The arrangement employed in France is known as the four a boulanger, or baker’s furnace.  The temperature attained in the furnace itself never exceeds low redness.  The material preferred is the softer kind of the granular variety of gypsum.  This is put in in pieces of about 21/2 inches in thickness.  After the baking several lumps are broken up and examined to see that there are no shining crystalline particles, which would indicate that some of the gypsum had remained unchanged.  Before use the plaster is ground very fine.  This point is of considerable practical importance.  The consistency attained should be such that the material may be rubbed between the finger and thumb without any feeling of grittiness.  Should there be particles of a size to be characterized as “grit,” these will after use appear at the surface of the mould, with the result that the mould will have to be abandoned long before it is really worn out, i.e., before the details have lost their sharpness.

It is manifestly of considerable practical importance to understand the conditions which determine the time of the setting up of plaster.  According to Payen, the rapidity of setting, provided the plaster has dehydrated at a temperature sufficiently low, depends entirely on the structure of gypsum employed.  Thus, according to him, the fibrous kinds gives a plaster setting almost instantaneously.  The water, he says, penetrates the material freely, setting takes places almost simultaneously throughout the mass.  The hydration of each particle is accompanied by an expansion, and under the conditions specified, this expansion being unresisted takes place to the maximum extent, with the result of leaving cavities between the crystals, and producing a set plaster of less coherence and density.  On the other hand, where granular crystalline gypsum has been used, setting begins at the surface of each group of crystals before the water has penetrated to the interior; the hydration is in consequence more gradual, and resistance being offered to the expansion of the inner parts, a harder and denser material is obtained.  That this expansion contains an element of truth is indicated by the practice of employing the granular crystalline variety for the preparation of moulding plaster.  The explanation appears, however, to be inadequate in several respects, especially in view of the fact that plasters for moulding are reduced to a fine state of division before use.  It seems as if this treatment must, in great part at any rate, break up the crystalline aggregates.

In order to discover a more satisfactory explanation, let us examine the results of the chemical analysis of plasters used in commerce.  One is struck by the large percentage of water they usually contain.  Thus, four samples of ordinary plaster analyzed by Landrin have an average of 90.17 per cent. of CaSO4 and 7.5 per cent. of water, while two samples of best plaster contained 89.8 per cent. of CaSO4 and 7.93 per cent. of water.  These numbers do not add up to 100, the difference being due to silica and other impurities of the original gypsum, amounting altogether to about 3 per cent.