The first attempt to imitate these natural forms would be made in wood and metal, substances which would lend themselves to the unpractised moulder more readily than stone, but in time the difficulties of the latter material would be overcome. The deep vertical grooves cut in the shaft would afford a rough imitation of the round stems of the lotus and the triangular ones of the papyrus. The circular belts at the top would suggest the cords by which they were tied to the shaft. The leaves and flowers painted upon the lowest part of the shaft and upon the capital, may be compared to permanent chromatic shadows of the bouquets of colour and verdure which had once hidden those members. Finally, the artist found in the swelling sides of the bud and the hollow curves of the corolla those flowing lines which he desired for the proper completion of his column. This hypothesis seems to leave no point unexplained, and it receives additional probability from a detail which can hardly be satisfactorily accounted for by the advocates of the rival theory. We mean the cube of stone which is interposed as a kind of abacus between the capital and the architrave. If we refer the general lines to those of a plain column bound about with flowering stalks, there is no difficulty. The abacus then represents the rigid column behind the decoration, raising its summit above the drooping heads of lotus and papyrus, and visibly doing its duty as a support. Its effect may not be very happy, but its _raison d'être_ is complete. On the other hand its existence is quite inexplicable, if we are to look upon the column as a reproduction in stone, a kind of petrifaction of a single stem. To what, in that case, does this heavy stone die correspond? To those who believe the capital to be the representation of a single flower with its circlet of graceful petals, its presence must seem nothing less than an outrage.