At all periods statues of royalty exhibit, however idealized, characteristic features that enable us to identify the subject. A unique collection of masks, evidently cast from living persons, was found in the studio of the sculptor Thutmose at Akhetaten. They evidently assisted the artist in making realistic or naturalistic portraits, but unfortunately few of these have survived. After a further phase of idealization the realistic tradition was fully re-established in the Late and Graeco-Roman Periods.

    The term kesty for sculptors also covered the carvers of stone and wood reliefs. The latter are shown on several fine reliefs chipping away with mallet and chisel on scenes already traced out by draughtsmen. Relief-carving was in fact one of the most frequent commissions given to artists. From the Old Kingdom up to the time of King Sethos I most temple and tomb reliefs were of the raised kind where the figures stood out, fully contoured, with the surrounding areas cut away. In the other, sunk reliefs, also represented in the Old Kingdom, the background is left uncut but the figures are carved in and beneath it. Sunk reliefs dominated temple walls from the time of Ramesses II, and exceptionally deep-cut reliefs are typical of the Ptolemaic Period.

    The distinction between draughtsmen and painters is reflected in the ancient Egyptian nomenclature. Draughtsmen are called sesh kedut, 'writers of outlines', showing the close affinity between drawing and writing. The old Egyptian script had, after all, evolved through the standardization of diagrammatic drawings, and scribe and draughtsman used the same instruments. The word for painter, sesh, denotes also a scribe.

    The activities of draughtsmen and of painters were closely associated. But as their pictures contain no information about their creative environment and methods, we have to rely on archaeological evidence and on partly-finished work. In addition to possessing originality and a flair for design, the ancient Egyptian artist needed to be fully conversant, not only with objects and events around him, but with various established and immutable religious preconceptions. These included the figures of the gods with all their attributes and the prescribed content of divine, ritual and royal scenes. But he was less bound by stereotypes when it came to portraying the lives of ordinary people.

    We get some idea of the artist's preliminary work from the ostraca used for practice by trainee draughtsmen and painters as well as by apprentice scribes. Even qualified craftsmen used them as cheap 'sketch pads' when preparing to work on the walls of tombs or temples, or to write on costly papyrus scrolls. These sketches furnish more testimony to the creative genius of artists, in fact, than do their final products, subject as these were to meticulous regulation of form and content. They often give a livelier rendering of movement - witness for example the picture of two cheetahs attacking an antelope on an ostracon in the Naprstek Museum in Prague, or the famous figure of a dancing-girl bending over backwards in the Egyptian Museum, Turin.

    On ostraca there are often sketched (cartoon-like) scenes which illustrate fables. There also occur ostraca with realistic preliminary sketches of human figures, even showing the use of perspective in their drawing, on which the final correction in black line reverts to the normal canonical style, to which we will return later.

    Another sketch-pad surrogate consisted of a little wooden board coated in stucco and marked out with a rectangular grid on which the artist made his drawing. In doing so he would adhere to the strict rules and then, having copied the grid onto a wall on a larger scale, transfer the design square by square.

    Use of a grid also ensured adherence to the basic rules of figure proportion that have been revealed by Erik Iversen and recently revised by Gay Robins. Up to the end of the Third Intermediate Period artists applied the 'first canon of proportion' based on the 'short cubit', that is the distance from the elbow to the tip of the thumb, conventionally set at 45cm. A human figure standing would be drawn onto a grid of 18 squares, each side of a square equaling the width of a clenched fist. Thus the length of a forearm was three squares, of a hand one-and-a-half and so on.

    The Saitic Period saw the introduction of the 'second canon of proportion' based on 21 squares. This had to do with the wider acceptance of the longer 'royal cubit' 52.36cm from elbow to tip of middle finger) which had previously been used only in architecture. The basic modulus, the width of a clenched fist, remained the same. So there were now three extra squares from top to foot of a human figure, of which one was assigned to the lower leg and two to the trunk, sometimes resulting in an unnatural elongation of the upper half of the body.

    In some cases the artist took the risk of sketching the figure straight on the wall-plaster while it was still wet, without a grid. An example occurs on the east wall of the South chapel in the 5th-dynasty tomb of Princess Khekeretnebty at Abusir, where the outline of a seated figure was drawn in white on the dark gray plaster. Usually, however, sketching was done in red, as we see in several scenes planned for the tomb of Horemheb in the Valley of the Kings (18th dynasty).

    The final drawing was then executed in a strong black line, ready for relief carving or coloring in. There are examples of this for instance, in other parts of Princess Khekeretnebty's tomb, in that of Horemheb and in the fine profile of a young princess in the I 8th-dynasty tomb of Kheruef at Asasif on the west bank of the Nile near Luxor. In addition to the canons of proportion there were other established conventions that the draughtsman had to follow. Successive scenes were arranged according to their content and prescribed order in 'registers', usually several one above another.