Cracks in plaster occur from several causes. If the distance between the ends of the laths, where they join on the studding or furring, is too great, the larger amount of plaster in that place, when drying out, may cause a short crack. Any such spaces should, however, be. filled by the lather before plastering is begun. Sometimes, too, especially in the first coats, cracks are caused by the shrinkage or expansion of the wooden laths after the mortar has wholly or partially set. The result is a series of narrow cracks parallel to each other and the width of the laths apart. Lath cracks are ordinarily filled in and covered up by later coats, and so do not often appear in the finished plastering They may, too, be worked out when floating up the coal before it finally sets. If wide or deep, however, they should be cut out to a width of an inch or so, and filled in with new mortar before adding the last coat.

Cracks of a like appearance are sometimes caused by the rough mortar being too rich, or by draughts of air from open dors or windows drying out portions of the plastering too quickly. The too rapid drying of plaster with stoves or salamanders, often produces a like result from similar causes. An experienced plasterer should be able to determine the responsible cause and take measures accordingly, using more sand if the mortar is too rich, screening openings to prevent draughts, and using less fire in his drying stoves. in green work, damage already done may be repaired by refloating again before the work becomes too dry, softening the mortar with water if necessary.

Cracks sometimes occur in the angles at the ceiling or corners of the room. When in this location, they may be caused by the shrinkage or settlement of the partition or floor. In the perpendicular angles, especially, they may extend only to the depth of the finishing coats. In that case, the causes are likely to be either too thick plaster, insufficient troweling, or an insufficient amount of plaster in the gauged coat - causes which are easily remedied in the remainder of the work.

Cracks running diagonally across a partition, or radiating from the corners of doors and window openings, are caused by the unequal settlement or shrinkage of the building. They frequently occur at a perpendicular angle where a wood partition is brought up against a brick wall, or at the ceiling line where a wooden floor comes up against a brick supporting wall.

Cracks occur in the final finish when the putty is not gauged enough or not troweled or brushed enough, when it is put on too thick, and when too little sand has been used. These cracks are called chipped cracks. Plaster, when apparently perfect and without cracks, will sometimes crumble, either from too rapid drying or from the use of too much sand. Either too much or too little sand materially injures the strength of mortar.

If unclean sand, dirt, or clay has become mixed with the mortar, it not only weakens the lime but prevents its adhesion to the sand particles, so that no real set of the mortar ever occurs. Of course, at all times, poor materials - sand, lime, or hair - may be responsible for defects in plastering. Plaster occasionally falls off even when apparently hard and good, if the laths are too near together, if there is insufficient hair, if the mortar is too rich or too sandy, or if it had not been pressed against the laths with sufficient force when being applied; or it may become loosened by the springing of the laths under the pressure of floating it too hard. On brickwork the mortar requires considerable more sand than for application on laths.

Lime must have time to set before it dries out. Therefore, to last well, it should dry slowly. A stiffer working mortar makes better and harder plaster than thin or wet material, provided, of course, it is thin enough to clinch well to the lath in first-coat work, or to adhere to brick and dry scratched surfaces, and to spread evenly, in second-coat work. Stiffer mortar can safely be applied upon wet mortar than on dry; and wide-spaced lathing will take stiffer mortar than close-laid laths. When two coats of mortar have been put on, and the last coat falls from the first, it is generally because the first coat was not wholly dry when the second was applied. The coats must either be entirely dry or quite green to be successfully combined.

If possible, it is better to have the workman use makes of materials, especially lime, having those properties with which he is acquainted. Attention has already been called to the fact that different makes of lime vary considerably in their chemical composition. It is not even certain that lime of the same make will always run even in production, year after year. Of course, lime that has been slaked by exposure to air or water while in the barrel, and before it is used, is worthless. As this occasionally happens, it is well to be watchful and see that such bad material is never added to the plaster bed.

As a final warning, be certain that the last coat of plaster has dried out hard and strong before any wood finish is installed, as otherwise the wood will absorb the moisture from the plaster, causing it to swell and therefore opening cracks that are never likely afterward to be altogether closed. All wood finish should also be kept out of the house while plastering is going on, as it will absorb moisture from the air around it. The reason that sash are not ordinarily set until after the plastering is finished, is because they absorb so much of the moisture as to cause the sash to swell in place. It is generally considered preferable to till the window openings or doors with screens of cotton cloth, as this prevents direct draghts and still allows of a circulation of air that dries plastering much more rapidly than artificial heat, or than it would dry if these openings were closed by solid doors and glazed sash. In very bad weather the screen of cotton may be slightly strengthened, if necessary, by the application of a coat of whitewash on the inner side. Contrary to what might be supposed, the cloth window-screen is almost as good a protection against external cold and frost as is the glazed window, although the current of air passing through the cloth meshes of these screens into and out of the house, causes a slight loss of heat, adding somewhat to the expense for fuel required to dry out a plastered building. In good drying weather, these screens should be taken out and left out during the day, but should be replaced at night or in damp weather, when the plaster otherwise is likely to reabsorb moisture from the air and so delay the time of its final drying out.

If avoidable, the artificial drying of plaster by salamanders should not be employed; natural drying by sun and air is. under all circumstances, preferable. The salamander not only dries the room in which it is placed, too quickly - especially the ceiling above - but fills the air and the plaster itself with gas fumes, and, by steaming, is frequently the cause of the rotting of plaster or hair, thus reducing its vitality and life. Heating a house to dry out the plaster by means of the regularly installed heating plant, is preferable to the use of salamanders, the chief objection in this case being occasioned by the unduly rapid drying-out of wall plaster back of or above registers and radiators.

The situation is helped if the radiator is set out from the wall and some screen is placed between it and the plaster. A screen may also be employed against the wall over a hot-air register, but there is no means of protecting the plaster on either side of partition through which a hot air or steam pipe passes. Such plaster is bound to be severely strained by being dried too quickly.

If plaster is frozen when wet, it is likely to loosen up and injure the whole mass so that it may eventually fall off. The effects of freezing are less troublesome if the wall is frozen after it is dried and has once set. If only slightly frosted, and thawed immediately and floated again, it may often be saved, the effect in that case being not much different from what it would be if the wall had been surface-moistened and refloated.