148 Primitive Greece: Mycenian Art. a single eruption, which may have lasted some time, but without notable break. A last observation will remove any lurking doubt Arguing from the fact that at the present day Thera has no potter's clay, and that it imports its earthenware from Milo and Anaphos, the multitudinous vases found within and without the prehistoric houses of Thera were at the outset held to have been brought to the island by way of trade. In order to be quite sure as to the origin of this pottery, MM. Fouque and de Cessac examined some fine slips of it under the microscope, and they discovered that the clay could have been supplied by the soil of Thera alone ; whilst the absence of anorthite lava further showed that the material was not derived from the north of the island, where this kind of felspar abounds, but from the south, near to the sea and fresh water, evidenced by the presence of diatoma, foramina, spongilla fluvialis, etc., in the clay of the vases ; that is to say animals, some of which live in salt and others in fresh water. No place answering to these conditions now exists ; but we learn from geology that a deep valley, before the great eruption, was inserted between the central cone and a hill fencing it in on the south. This valley came up close to the islet of Aspronisi, and from the deposit formed at the mouth of a stream which flowed at the bottom of the valley came the clay of our vases.^ It follows therefore that they were made at Thera itself before the catastrophe. Moreover the excavations have proved that the structural block described above did not stand alone, but that veritable hamlets rose on several points of the island, suggesting a thick population. Thus, the dwelling of which we give a plan contained apartments other than those that have been unearthed. Connected with the northern wall was a second and lower one, stretching towards the west below the tufa, and bounding a great vaulted chamber, which was only partially emptied, to guard against a sudden collapse of the walls enclosing it. Here too, on the same level, they traced other walls ; whilst a smaller erection consisting of a single room, and the beginning of a wall which presently disappears under the cliff, were uncovered at about twenty-four metres from the north-east corner of the enclosure. It was not only at Therasia that this population left traces of its activity ; we also find unequivocal marks of their activity ^ FouQufe, Santorin.
