It has long been a familiar fact that gypsum yields on baking a material which possesses the power of setting with water to a firm mass, this setting being accomplished much more quickly than is the case with mortar.

The explanation of the setting of plaster was first given by Lavoisier, who pointed out that gypsum is an hydrated salt, and that the set plaster is in fact gypsum reformed, the change brought about by baking being merely loss of water of crystallization. The beds of gypsum of most importance both formerly and at the present time in the plaster manufacture occur in the neighborhood of Paris in the lower tertiary formation. Different beds differ (1) in respect of character and quantity of admixed materials and (2) in the structure of the gypsum itself. With regard to the first point, some deposits contain a notable proportion of carbonate of lime, a fact which under certain circumstances may considerably influence the character of the plaster. In the matter of structure two principal varieties occur (1) granular and (2) fibrous. Further, hardness of the granular kind varies considerably. These differences of structure in the original material appear to exercise an influence on the properties of the plaster. Thus according to Payen the plaster formed from the granular variety sets more gradually than that derived from the fibrous, and forms a denser mass.

The softer kinds of the granular gypsum are those principally used in the production of plaster for the moulds of potteries.

In the old fashioned process which is still employed for making the common kinds of plaster, the material is exposed to the direct action of flame. Large lumps are placed in the lower part of the furnace, above them smaller lumps, and, after the heating has been carried on for some time, finely divided material is filled in at the top. The outer portion of the larger lumps is always overburnt, and in the upper part of the furnace the presence of shining crystalline particles generally indicates the fact that some gypsum has remained unchanged. Provided that the amount of unburnt and overburnt material does not exceed about 30 per cent. of the total, the plaster is suitable for many applications.

It was early observed that set plaster could be revivified by a second baking, but attempts in this direction were not uniformly successful, it being found that the dehydrated substance in some cases refused to set with water. It behaved in fact similarly to the natural anhydrous calcium sulphate which is unaffected by water. These failures were found to be due to the employment of too high a temperature, and such plaster was termed dead burnt. Although this fact was ascertained long ago, yet ignorance of what had already been done has probably been the cause of many disappointments in attempts at revivification which have been made from time to time by persons unacquainted with the history of the subject.

The view generally adopted with regard to the theory of these processes is that plaster consists of anhydrous calcium sulphate, CaSO,in a condition probably amorphous, different from that of natural crystallized CaSO, known to mineralogists under the name of anhydrite. By the influence of a high temperature it appears probable that a molecular change is gradually induced with production of a crystalline structure, and probably an increase of specific gravity, resulting in the artificial reproduction of the mineral anhydrite. No determination appears to have been published of the specific gravity of plaster prepared by complete baking at a low temperature. The theory is, however, confirmed by the results obtained by workers on the subject of mineralogical synthesis, who have shown that the material which has been produced at high temperatures has the specific gravity and other physical properties of the mineral anhydrite.

It was formerly supposed that plaster prepared by baking at a temperature above 300 degrees loses completely its power of setting. Later observations, however, as those of Landrin, negative this view. Between 300 degrees and 400 degrees Landrin obtained plasters setting almost instantaneously when mixed with a small amount of water. When the temperature employed approached 400 degrees, the set plaster was softer, but the setting still took place quickly. These observations appear to show that the change to anhydrite is a very gradual process at temperatures below a red heat.

Reference has been made to the differences in (1) time of setting of plaster and (2) in hardness of the resulting material. Both of these properties are affected by the mode of baking. The hardest material is frequently obtained from the quick-setting plasters, but for certain purposes this rapidity in setting is of great practical inconvenience. Thus the moulder in pottery work must have leisure to fill in every detail of a design often complicated and intricate before the material with which he is working becomes intractable. Thus for many of the more refined purposes to which plaster is applied, extreme hardness in the set plaster is of less vital importance than a convenient period of setting. On the other hand, plasters which set very slowly give as a rule too soft a material, as well as being inconvenient in use. Plasters which hit off the happy medium are alone suitable for the work of the potter. The finer varieties of plaster prepared especially for use in potteries are obtained by a treatment which differs in many respects from that described above for the commoner kinds. In the first place, the direct contact of fuel or even flame is avoided, since this reduces some of the sulphate to sulphide of calcium, the presence of which is in many respects objectionable.

Secondly, it is necessary that there should be a better control over the temperature, since, as has been seen, if the heating be carried too far the plaster, if not partially dead burnt, will set too quickly for the particular purpose to which it is to be put.