It was Minoan writing that led Arthur Evans to Crete. In Greek antique shops and elsewhere he had seen certain stone signets, each with three or four facets inscribed with what he felt sure were characters of some prehistoric system of writing (cf. Suppl. PL CLXXVII (a)); they were alleged to come from Crete. After a visit to that island in 1894 to collect further evidence he was able to recognize, besides these ‘hieroglyphs’, a later and more linear type of writing occasionally scratched on vessels of pottery or stone from prehistoric sites in Greece and Crete. In a long article in the Journal of Hellenic Studies for 1894 he hazarded some remarkable conclusions from this scanty evidence: that although the ‘hieroglyphs’ probably concealed a pre-Greek language, the later, linear script might have been used for writing Greek; and that this linear script was a syllabary somehow related to the syllabary used in Cyprus in historical times as an alternative to the Greek alphabet. He was right on both points; but paradoxically his excavations at Knossos, begun in 1900 in pursuit of the origins and history of these scripts, led him gradually farther from the truth. Within the first week of digging he had begun to unearth clay tablets written in what he later dubbed ‘Linear B’ script (cf. Suppl. Pis. CLXXVIII (a), (b), CLXXIX and CLXXX (a)); and they were associated with objects at once identified as belonging to the Mycenaean civilization known from mainland Greece since Schliemann's excavations in the 1870'8. Further exploration, however, revealing the earlier stages of the civilization of Knossos, made it necessary to distinguish a distinct ‘Minoan’ civilization, in many ways a parent of the Mycenaean. Evans saw this civilization more and more as continuous, and as non-Greek. The corollary of this would be that the hieroglyphic script and the two linear scripts that succeeded it (for he now recognized an intermediate stage—‘Linear A’ (cf. Suppl. PI. CLXXVII (cj)—as well as the Linear B of his Knossos tablets)—all represented one non-Greek ‘Minoan’ language.