PAINTING MATERIALS
in Persian Miniatures/ p. 146) that blue from lapis lazuli was used in Byzantine
illuminated manuscripts from the VII century on but that the quality of the early
blue was poor. Laurie adds (he. cit., p. 148) that this dull blue was used in Persian
miniatures in the XIII and XIV centuries, but that in the XV century it was re-
placed by fine ultramarine which may have had its source in the East but was
refined in Europe and then returned to the East again. In China, however, azurite,
not ultramarine, was the chief mineral blue.
Methods for purifying and concentrating ultramarine from the crude lapis
lazuli appear to have been developed in the West in the XII and XIII centuries,
although the raw material still came from the East. These methods have been
handed down in numerous recipes which are quite similar in principle and vary
only in details. One of the best is given by Cennino Cennini (see Thompson, The
Craftsman's Handbook, pp. 36-39), who directs that the powdered mineral be
kneaded in a weak lye solution with a paste or dough of wax, pine rosin, linseed
oil, and gum mastic. The dough retains the foreign particles (silica, calcite, pyrite,
etc.), but the fine particles of blue color settle out in the alkaline water. The first
extraction gives the finest and purest color; each successive extraction gives a
product less pure until, finally, there is a pale gray-blue called * ultramarine ash/
The reason for the separation is the preferential wetting and retention of the
impurities by the dough mixture. Microscopically, natural ultramarine is charac-
teristic in appearance, and one can quite easily distinguish it from modern
artificial ultramarine (see Ultramarine Blue, artificial). The particles are clear
blue and translucent; their fracture is conchoidal and, when not too finely divided,
certain ones with squared corners and others shaped like sharp splinters can be
seen. The blue particles are isotropic and have a very low refractive index,
n = 1.50, lower than dried linseed oil or Canada balsam. Unless it is very highly
purified, ultramarine contains colorless, birefracting particles of calcite; occa-
sionally small, golden particles of iron pyrites can be seen by reflected light. Since
the refractive index of ultramarine is so low, it served better and was far brighter
in tempera than in oil. It has apparently discolored (turned green) in many old
paintings because of the yellowing of oil and varnish films that are applied over it.
Natural ultramarine is unaffected by red heat or by alkalis but is decomposed
by dilute acids, even acetic acid, with complete loss of color and the discharge of
hydrogen sulphide gas (see Ultramarine Blue, artificial). This sensitivity to acids
•may be the cause of the so-called e ultramarine sickness/ a phenomenon that is
occasionally met with in old pictures where areas painted with ultramarine have
turned gray-blue. As explained by De Wild (pp. 14-16), this may perhaps be
caused by either an acid medium or varnish or by an acid atmosphere. The blue
is stable, however, in strong light and many specimens which are several hundred
years old show no apparent change in color.
In mediaeval times, natural ultramarine was a costly material; it was in a
class with gold as a symbol of luxury and it was frequently specified by the rich