edifice then erected at Rome, were probably always taken from older buildings, and the entablature frequently made out of the fragments of former entablatures; but in the portico of that church there is an entablature made for the building, with a high frieze in mosaic, and the mouldings of the cornice are composed of lines nearly straight, and with ornaments of little relief. Of buildings erected during the government of the empress Placidia, and afterwards under the reign of Theodoric, that is, from 425 to 450, and again in the first half of the sixth century, there are seven churches, a baptistery, and two mausolea, existing at Ravenna, which I have already described. In the time of Dioclesian, and before that period, when arches and columns were used together, the arches sprung from the top of the entablature; in Constantine’s time, the practice was to spring them from the capitals of the columns; and in some countries we find this practice existing to a comparatively recent period, especially in cloisters. But the northern nations, in their attempt to copy Roman architecture, imitated rather that of the time of Dioclesian, than of a later date. They however, diminished very much the entablature, and ultimately reduced it to a mere slab over the capital, on which some of the appropriate ornaments of the ancient entablature might still be traced. On the other hand, the architects employed by Theodoric introduced a solid block under the springing, which is evidently derived from the construction of a stone arch, and not from that of a wooden entablature. The same arrangement occurs at St. Mark’s at Venice. This peculiarity forms one striking point of difference between the architecture of the age of Constantine, and of that of Theodoric; another is in the use of corbels: the small columns of the latter age frequently standing upon them; and the impost of the arch is sometimes lengthened out into a corbel, and supported by a column, which is not placed under its extremity; the impost continuing beyond its support, and sustaining a wall, considerably advanced before the face of the columns. Something of this sort occurs in Dioclesian’s palace at Spalatro, but I have not observed it in any of Constantine’s buildings.