GRECIAN AND MODERN VILLA FURNITURE. 1948 106'^ are in good taste, let us first see h w far they display unity of style. In the legs and bottom rails this unity is con- spicuous. Looking at the former apart from the latter, we should never expect to find the bottom rails joined to them, square in the section ; neither should we expect cylindrical rails perfectly plain. The legs being turned and carved, the rails, to be in unison with them, ought to be turned and carved also ; and, as they are so, they are therefore in good taste. The style of the supports of the table ought, of course, to give a general idea of the style of the top, and here we find that some of the tops, for example those of figs. 1949 and 1951, have carved mouldings round the edges, and therefore they harmonise with the legs and rails, and consequently may be considered in good taste ; but the tables, fig. 1948, fig. 1950, and fig. 1952, though their legs are equally en- riched with those of the other De- signs, have plain mouldings on their edges ; hence we have no hesitation in determining them to be in comparatively bad taste. The justness of this criticism will appear more obvious, by applying it to the table, fig. 1948, which has two supports more highly en- riched by carving than those of any of the other figures, and yet has a plain top. This is in bad taste, and ought not, in a work like the preser.t, to be passed over without notice. In order to show the utility of this kind of criticism, and of giving Designs which are less perfect than they might be made, let us suppose, for a moment, that we had only given the tables figs. 1949 and 1951, which we allow to be in tolerably good taste. All that we could have said of these tables is, that they were very good of their kind, displaying unity of style and of mLllLiLJL
