When engraving on hard stone was first attempted, it was, then, less for the love of art than for the profit to be won by the magic virtues and mysterious affinities, both of the material itself, and of the image cut in its substance. Then, with the increase of material comfort, and the development of social relations, came the desire of every Chaldæan to possess a seal of his own, a signet that should distinguish him from his contemporaries and be his own peculiar property, the permanent symbol of his own person and will. So far as we can tell, none but the lowest classes were without their seals; these latter when they were parties or witnesses to a contract, were contented with impressing their fingernails on the soft clay. Such marks may be found on more than one terra-cotta document; they answer to the cross with which our own uneducated classes supply the place of a signature. When the use of the seal became general, efforts were made to add to its convenience. In order to get a good impression it was necessary that the design should be cut on a fairly even and regular surface. The river pebbles were mostly ovoid in form and could easily be made cylindrical by friction, and the latter shape at last became so universal that these little objects are always known as _cylinders_. These cylinders were long neglected, but within the last few years they have been the subject of some curious researches.[296] They may be studied from two different points of view. We may either give our attention to the inscriptions cut upon them and to their general historical significance, or we may endeavour to learn what they may have to teach as to the religious myths and beliefs of Chaldæa. As for us we are interested in them chiefly as works of art. It will be our duty to give some idea of the artistic value of the figures they bear, and to describe the process by which the engraving was carried out. The cylinders are, as a rule, from two to three-fifths of an inch in diameter, and from three quarters of an inch to an inch and a half in length. Some are as much as an inch and three quarters, or even two inches long, but they are quite exceptional.[297] The two ends are always quite plain--the engraving is confined to the convex surface. As a rule the latter is parallel to the axis, but in some cases it is hollowed in such a fashion that the diameter of the cylinder is greater at the ends than in the middle (Fig. 131). Nearly every cylinder is pierced lengthwise, a narrow hole going right through it. Those that have been found without this hole are so very few in number that we may look upon them as unfinished. In some cases the hole has been commenced at both ends, but the drill has stopped short of the centre, which still remains solid. [Illustration: FIG. 131.--Concave-faced cylinder; from Soldi.] [Illustration: FIG. 132.--Cylinder with modern mount; from Rawlinson.]