Certain chambers at Knossos, paved and lined with gypsum, and two in Melos, have square central piers. These seem to have had a religious significance, and are possibly shrines devoted to pillar-worship. The houses of the great dead were hardly less elaborate. The “Treasury of Atreus” had a moulded façade with engaged columns in a sort of proto-Doric order and marble facing; and there is good reason to suppose that its magnificent vault was lined within with metal ornament or hanging draperies. The construction itself of this and the other masonry domes bespeaks skill of a high order. For lesser folk beehive excavations were made in the rock, and at the latest period a return was made apparently to the tetragonal chamber; but now it has a pitched or vaulted roof, and generally a short passage of approach whose walls converge overhead towards a pointed arch but do not actually meet. The corpses are laid on the floor, neither mummified nor cremated; but in certain cases they were possibly mutilated and “scarified,” and the limbs were then enclosed in chest urns. There is evidence for this both in Crete and Sicily. But the order of burial, which first made Mycenæan civilisation known to the modern world, continues singular. Similar shaft graves, whether contained within a circle of slabs or not, have never been found again. The latest excavation has at last established beyond all cavil that the civilisation which was capable of such splendid artistic achievement was not without a system of written communication. Thousands of clay tablets (many being evidently labels) and a few inscriptions on pottery from the palace at Knossos have confirmed Mr. A. J. Evans’ previous deduction, based on gems, masons’ and potters’ marks, and one short inscription on stone found in the Dictæan cave, that more than one script was in use in the period. Most of the Knossos tablets are written in an upright linear alphabetic or syllabic character, often with the addition of ideographs, and showing an intelligible system of decimal numeration. Since many of the same characters have been found in use as potters’ marks on sherds in Melos, which are of earlier date than the Mycenæan period, the later civilisation cannot be credited with their invention. Other clay objects found at Knossos, as well as gems from the east of Crete, show a different system more strictly pictographic. This seems native to the island, and to have survived almost to historic times; but the origin of the linear system is more doubtful. No such tablets or sealings have yet been found outside Crete, and their writing remains undeciphered. The affinities of the linear script seem to be with the Asianic systems, Cypriote and Hittite, and perhaps with later Greek. The characters are obviously not derived from the Phœnician.