150 Primitive Greece: Mycenian Art. specimen of metal which these ruins have disclosed ; ^ one of its ends is pointed, to facilitate insertion in the wooden handle (Fig. 31). Despite the poor state of the teeth and its greenish colour, which is due to carbonate of copper, it is still sharp enough to cut soft wood, such as pine and poplar. M. Gorceix, basing his conclusions on the fact that the pumice filling the apartment where lay the copper instrument was undis- turbed, and that the stratifications when visible along the cliff are manifestly regular and horizontal, — which could scarcely be the case had there been a partial subsidence, that would have buried the houses and carried the tool below, — rules it to be contem- poraneous with the houses. Mixed with a thick bed of straw, strewing the floor of the apartment wherein the saw under notice was discovered, were countless goat or sheep bones, and earthen bowls which contained barley. The excavators found on other spots bones of the same nature, showing that the men who Fig. 31. — Copper saw, after Gorceix. Length^22 cent, by 5. inhabited these villages had domestic animals, whilst the disks with central holes referred to above indicate no less clearly that the fish which abound in these waters were caught with nets which the whorls in question weighted and kept stretched (Fig. 32). From the variety of grains contained in huge vases, such as lentils, aniseed, coriander, and a kind of pea known at the present day in the island of Thera as '* arakas, and barley above all, we may conjecture that husbandry had made considerable advance. Nor was tree culture unknown : at least there is evidence that the hilly slopes were already shaded with the graceful foliage of the olive ; for amidst these ruins have been found branches and stems of this tree, some of which , still pre- serve their bark. Did they know how to graft it, and press out the oil from its fruits ? To this query we have no very clear answer to give. But we might be tempted to think they did, ^ For a circumstantial account of the question, the reader is referred to GorVeix, Bulletin de P Ecole fran^aise (PAthines^ 1870).
