But it is futile to insist upon any such comparison. The polygonal column had long been disused when the Greeks first penetrated into the Nile valley and had an opportunity of imitating the works of the Egyptians. It was in use in the time of the Middle Empire, during the eleventh and twelfth dynasties. The earlier princes of the Second Theban Empire introduced it into their stone buildings, but there are no examples which we can affirm to be later than the eighteenth dynasty. The Rameses and their successors preferred forms less bold and severe; their columns were true columns with swelling entasis and rich and varied capitals. It is no doubt true that towards the seventh century the Greeks could find the polygonal column which we have described in many an ancient monument. But those early visitors were not archæologists. Astonished and dazzled by the pompous buildings of a Psemethek or an Amasis, they were not likely to waste their attention upon an abandoned and obsolete type. Their admiration would be reserved for the great edifices of the nineteenth and later dynasties, for such creations as Medinet-Abou, the Ramesseum, and the Great Hall at Karnak; creations which had their equals in those cities of the Delta which were visited by Herodotus and Hecatæus. If Greek art had borrowed from the Egypt of that day it would have transferred to its own home not the simple lines of the porticos at Beni-Hassan, but something ornate and complex, like the order of the small temple of Nectanebo at Philæ. These few words had to be given, in passing, to an hypothesis which has found much favour since the days of Champollion, but we hasten to resume our methodical analysis of the Egyptian orders, and to class them by the varieties of their proportions and by the ever-increasing complication of their ornaments. [Illustration: FIG. 74.--Polygonal column with a flat vertical band.] [Illustration: FIG. 75.--Polygonal pier with mask of Hathor; from Lepsius.] At Beni-Hassan and elsewhere we find pillars with two or four flat vertical bands dividing their flutes into as many groups. These bands are covered with incised inscriptions. Sometimes, as at Kalabché (Fig. 74), there are four flat bands inclosing five flutes between each pair. Such an arrangement accentuates the difference between these so-called proto-doric pillars and the Greek doric column. They take away from the proper character of the pillar, the inscribed tablet becomes the most important member of the composition, and the shaft to which it is attached seems to have been made for its display. In the Greek order, on the other hand, we always find the structural requirements brought into absolute harmony with those of the æsthetic sentiment; every line of every detail is necessary both to builder and artist.