Restoration and Preserva Tion of Paintings

What is needed for de-varnishing is not a rapid solvent, but something which will soften the varnish gradually. It is necessary to keep the operation well under control the whole time, and not to use a method which gets out of hand. The work must be slow if it is to be safe. A Van Goyen, painted in transparent browns applied directly on wood, a Reynolds, the colour of which is sometimes applied simply by scumbling on a ground of grisaille, or a Debucourt, are extremely difficult to de-varnish properly. Excellent and sure results can however be obtained if we have patience to attack the varnish not all at once (as is done by alcohol, alkali, soap containing potash, or acetic acid applied without precaution) but gradually. This makes it possible to modify the process in the middle of the operation when one coat ing of varnish has been removed, and we arrive at another coat ing which may be of an entirely different composition. For it may be 'taken almost as an axiom that all pictures have been varnished more than once.

Many people advocate the rubbing off of varnish. It is true that certain varnishes can be rubbed off by the fingers, and that this method can be used for de-varnishing small pictures painted on a smooth surface; but it is defective and irregular, if not impossible, if there is much impasto. On the other hand, oil of aspic, copaiba balsam, cedar oil, sulphuric ether and xylol may be used successfully either separately or in combination ; they have the advantage of allowing a trained expert to recognize the na ture of the varnish which he has to remove and to choose the most appropriate formula for his work. If judiciously used alkali, acetone and acetic acid which have been mentioned as dangerous may be of great value. It must be emphasized, however, that de-varnishing ought to be perfectly successful and perfectly safe. Different methods should be used according to the nature of the picture and the varnish with which it is covered, for in this work each case should be considered individually. If there is nothing wrong with the picture except that the varnish is too dark it is sufficient to revarnish it lightly after the old varnish has been removed. It is, however, better to wait before doing this for a sunny day so as to be sure that no moisture is retained between the paint and the varnish.

Restoration.

It is now necessary to consider the case of pictures which have to be restored after the old varnish has been removed. It very rarely happens that an old picture has not suf fered from scratching, if from nothing worse. The preceding paragraphs have dealt with re-backing and transference ; these processes presuppose some injury to the painting itself which has to be repaired.

It would take too long to distinguish between all the various kinds of damage which may necessitate the restoration of a pic ture, that is to say the reconstitution of the painting itself. For

the sake of clearness they may be divided into two categories : (I) absence of paint, which may be due to a hole, a tear, or to scaling off ; (2) wearing of paint, which may be due to changes occurring in the paint itself ; e.g., when a red or brown ground works through scumbling which has been applied over it too lightly, or to friction by foreign bodies, or, what is unfortunately the most frequent cause, to previous cleaning operations which have worn away lightly painted details or damaged fragile colours, such as scumbling on the one hand and all dark tones on the other.

Retouching.

In addition to the above two kinds of injury, there is another factor of great importance—that of old re touchings. By "re-touchings" we understand restorations which go beyond the area of the injury which they were intended to repair. When this has been done, the work of restoration is rendered doubly difficult, because if the work is to be properly done it is first of all necessary to remove all the additions which have been made to the original painting. A trained eye can generally detect re-touchings at once, and can guess whether they have gone beyond the actual area of the accident.

Ultra-violet rays and X-rays are of great service in this con nection. Under the action of ultra-violet rays, colours, at any rate if not covered with varnish, are transformed in accordance with their chemical composition, and it may happen (though it is not safe to generalise) that "values," or colours which closely resemble one another by natural light, are completely differen tiated by ultra-violet rays. If the paint used for re-touching is not of the same chemical composition as that of the original picture, it will at once appear under ultra-violet rays as a patch of a different colour.

X-rays perform a similar service for a different reason. It is well known that the greater the atomic weight of a body, the more difficult it is for X-rays to pass through it. (See X-RAY.) Consequently a re-touched area, which represents a greater body of material on one part of a picture, is shown up by the fluorescent screen. Here again it is unsafe to generalize, for certain colours, such as browns, emerald green and madder, are of such low atomic weight that they are scarcely shown by X-rays. On the other hand, X-rays reveal the exact limits of an injury. When we see that a painting has been retouched, or even, as sometimes hap pens, completely painted over, we may hope that the real extent of the injury is not so great as the re-touching would lead us to suppose. By. the use of X-rays it is generally possible to dis cover the exact area of the mastic underneath the retouching, as the atomic weight of the mastic is sure to be different from that of the wood or canvas on which the picture is painted.