440 MERGE TO GYRENE.

There were, however, very few instances in which the established laws of proportion, so far as propriety and apparent security are Above the capital of the triglyphs, between it and the cymatium below the corona, there is usually a band or fillet, of the samede23th, for the most part, with the capital, and on the same plane with it ; and the capital itself sometimes projects a little beyond the femora of the triglyph, and sometimes is on the same level with it. The cymatium below the corona is for the most part much deeper than the usual proportion of that member ; which appears to have been done in order to show the ornament upon it, which would not otherwise, from the projection of the corona and the depth of the mutules, be conspicuous. The proportion of the corona itself also varies, and the scotia beneath it is sometimes introduced, and sometimes omitted altogether. Much difference exists in the depth of the cyma, as well as in that of its fastigiura ; and the lions’ heads, which are often sculptured upon it, are sometimes introduced and sometimes omitted. Whenever these are placed, as they usually are, over the axes of the columns, an ornament representing the end of a tile is often found to accompany them, placed on the fastigium, exactly over the centre of the metopes. There is also a difference in the depth of the regulse and mutules, as well as in the thickness and depth of the guttse, the form of the latter being sometimes conical and sometimes cylindrical, and on some occasions almost square. The upper part of the two outer channels of the triglyphs are sometimes cut parallel with the line of the capital ; but more frequently inclined a little downwards, so as to meet the bottom of the moulding above the metopes, which we have already stated is not often in a line with that of the capitals of the triglyphs. The depth of the taenia, also, and that of the epistylium (or architrave) varies in different instances ; as well as the proportions of the columns themselves, and those of their abaci, or plinths : the latter are generally surmounted with an elegantly proportioned cymatium, which is itself almost always crowned with a fillet. We may add that the curve of the echinus also varies, but is usually of a light and elegant proportion ; and the annulets sometimes follow the line of the curve, and sometimes range with that of the hypotrachelium : the number of these occasionally two, but more frequently three ; and the upper and lower ones (in the last-mentioned instance) are frequently cut square, while the central one forms an angle, the apex of which projects beyond the two others. This, however, only occurs when the annulets range with the shafts of the columns, for when they range with the line of the echinus they are generally cut like the teeth of a saw, as the central one is in the instance just mentioned. We observed that for the most part when annulets were adopted there were no channels, or grooves, hollowed in the hypoti'achelium, and this equally obtained whether the annulets followed the line of the echinus or that of the