During the Ptolemaic period, the Egyptian architects made frequent use of the form of capital which is now called _hathoric_, in which a masque of Hathor, the cow-headed goddess, is the ruling principle. This capital is to be seen, in a rudimentary condition, in a pavilion dating from the fifth dynasty (Figs. 60 and 61). It there occurs, as will be seen by referring to our illustrations, as the roughly blocked-out head of a cow. In connection with the last two bas-reliefs, we must call attention to the fact that the structures from which they were imitated must have been erected in some kind of metal. Their forms are inconsistent with the use of any other material. The way in which the capital is connected with the member to which it acts as support, in Fig. 59, and the open-work of the architrave in Fig. 61, are especially suggestive. In the latter bas-relief the figures introduced are evidently behind a grille, and the whole structure is expressive of metal-work. [Illustration: FIG. 60.--Bas-relief from the 5th dynasty; from Lepsius.] [Illustration: FIG. 61.--Details of the columns.] We suspect that the pavilion shown in Fig. 56 was also of metal, which seems to have played an important part in all that light form of architecture with which we make acquaintance in the sepulchral decorations. This is very clearly seen in the examples of painted columns, which we borrow from Prisse (Figs. 62-65). They present forms which could only have been compassed by the use of some metal like bronze. If the use of metal be admitted, we have no difficulty in accounting for the playful and slender grace found in some of these columns, and the ample tufted capitals of others. The natural tendency in painted decorations of this kind to exaggerate the characteristics of their models must not, however, be overlooked. Not being compelled to apportion the strength of supports to the weight which they have to carry, it is always inclined to elongate forms. The decorations at Pompeii are a striking instance of this. Pompeian painters gave impossible proportions to their columns, which evidently existed no where but in their own fancies. We admit that the Egyptian decorators did something of the same kind, that they exaggerated proportions and accumulated motives on a single capital, which were not to be found co-existing in reality. But, with these reserves, we think it more than probable that the columns shown in their paintings have preserved the general aspect of the supports employed in those curiously elegant pavilions to which they belonged. The forms in Fig. 62 are explained, on the one hand, by the imitation of vegetable forms, on the other by the behaviour of a metal plate under the hand of the workman. The curve which was afterwards, under the name of a _volute_, to play such an important part in Greek architecture, was thus naturally obtained. [Illustration: FIGS. 62-65.--Columns from bas-reliefs (Prisse).]