344 PROCEEDINGS OF THE AMERICAN ACADEMY 



barrel.* This method of preventing the bursting of the column under 

 its burden is noticeable, because it confirms the theory (presently to be 

 enunciated) of the source of this columnar form, and because it accords 

 with the explanation just offered of the expansion of the capital. For 

 since an incumbent weight, sufficient to break down and bend back 

 upon themselves the weaker tops of the stalks, would tend to spread 

 apart the stiffer parts below, the presence of these bands implies the 

 existence of such a pressure. In many of these twice or thrice bound 

 columns the number of constituent stalks is different in different sec- 

 tions of the shaft. This may result from a fancied insertion of smaller 

 stems in the intervals of the larger, but is more probably an indepen- 

 dent variation of the type. 



Lepsius gives (i. 107) an engraving of one Ptolemaic column from 

 Philre, whose capital is composed of several lotus-buds ; but it must 

 be regarded merely as a capricious deviation from the antique type, 

 which needs no extended description. 



It only remains to trace as best we may the genesis of these forms 

 from natural types.f The possibility of thus tracing back its members 

 to their origin, is one of the peculiarities of Egyptian architecture. 

 Its types are not buried, as are the Greek and those of all subsequent 

 styles, under many successive layers of alteration and refinement, but 

 lie close to the surface, ready to be uncovered by any investigator. 

 The earliest and simplest architectural method is here clearly exempli- 

 fied in the free use of natural and purely mechanical forms as stepping- 

 stones to artistic products. Of this method this group of columns is 

 an excellent representative. 



Sir Gardiner Wilkinson, as already remarked, regards these bundle 

 columns as resulting from the decoration of square pillars.} In refuta- 

 tion of this view, it may be urged that, while utterly insufficient to 

 explain the papyrus forms, it is highly improbable for the lotus group 

 to which it was intended especially to apply, for two reasons. In the 

 first place, there are no satisfactory illustrations of the progress of 

 development from the unadorned pier to the comely shafts with their 

 slender stems and buds.§ On the one hand, we have piers in the 



* E. g., Gwilt, Diet of Architecture, p. 37. 



t Type, as I have used the term, is a form, whether natural or mechanical, 

 which occasions the conception of an architectural form hitherto unemployed. 



t Kg. in Time of Pharaohs, p. 163. 



§ This refers of course to the indications of a gradual development of concep- 

 tions. No one could justly claim that a full series of actual remains is necessary 



