234 Primitive Greece : Mycenian Art. the gaps between them being filled with smaller units (Fig. 62). A little further we come upon more regular masonry, but the interstices are still made good with pebbles {Fig. 63), whilst elsewhere there is a marked tendency to horizontal beds. The Fin. 6z.~WaU on the Bali Dagh. Stones are about forty centimetres in length and fifteen centimetres in height. As a rule, each successive course is set back from the one immediately below (Fig. 64). Are we to recc^nize i,n these walls the outcome of two distinct epochs, or a mere difference of execution sufficiently accounted for by the assumption that two sets of workmen were employed, and that each worked after its own habitual fashion ? To this question no satisfactory answer can be given. Irr^ular masonry is not sufficient evidence for fixing the date of a wall ; both Greece and luly offer numerous examples of polygonal masonry long after the