inserted vowels, rather than the correct nonvocalic forms. The name hemhem is used for a crown made of three reed bundles with disks on their tops, a feather at each side of the group, and a pair of horns supporting the whole; a fine example is shown in Daressy's Statues des diνinités, Pl. 11, Νo. 38201. I do not know on what authority the name rests; it is used by Poole, British Museum Catalogue of Alexandrian Coins (p. lxv and at Νos. 902, 1130, 2213), and I have seen it elsewhere.
“Ouroboros” is the name given by the Greeks to the design of a snake devouring its own tail. To some late writers it symbolized the universe, to others recurring time or simply the year. It occurred rarely in dynastic art, and is thought to be a substitute for the common representation of the Ηeaven goddess making an arch over the Εarth, which closes it below, completing the circuit (Horapollo 1, 2, with Sbordone's note). Though it is extremely common on magical amulets, there is nothing in the designs and inscriptions that occur with it to suggest a definite meaning for it; it seems to have become little more than a conventional border for such stones.
Archaeologists commonly use the word “cynocephalus” in referring to the “dog-headed” baboon, which Εgyptian belief associated with the god Thoth as his attendant and sometimes as his representative. It was also believed to worship the sun at its rising.
With a few exceptions, which are mentioned as they occur, the illustrations are of the same size as the objects represented. The longest measurement of an object is given first without regard to its shape; hence the terms “upright oval (or oblong οr rectangle)” and “transverse oval” have been found convenient, the former meaning that the vertical axis of the design lies in the longer dimension of the surface, the latter that it is in the shorter. I have noted the thickness of all amulets that I have examined in recent years, except metallic pendants and medals, which vary little from one millimeter. But when the object had to be described from a cast, that dimension is not given; and not only thickness but length and breadth also are only approximate when stones are mounted in a setting. All measurements are given in millimeters, and fractions of millimeters are disregarded or else taken to the next higher unit when there is reason to think that the edges have worn down.
The photographs used in making the illustrations generally represent plaster casts (not impressions) of the objects. Direct photographs of gem stones are less satisfactory, especially when the object is translucent, when it has a highly polished surface, reflecting light, and when it is of a dark color or of more than one color, as in banded or mottled stones. For this reason stones were photographed directly only when casts could not be obtained time to be used, or occasionally as an experiment. In the bronze objects the practice varies, direct photographs being more commonly used for pendants and medals, though there are exceptions which will be noted.
Casts have been preferred to impressions because they reproduce the designs as seen on the original stones. All students of magical gems have observed that, except in a very few specimens, the design was meant to be