OF ARTS AND SCIENCES. 349 



"With reference to the idea involved in the body of the shaft, it is 

 easier to say what it is not than what it is. The shaft is obviously 

 not understood to be composed of separate stalks, like the bundle 

 columns. But whether a single round stalk is symbolized, or the 

 superficial shape of a bundle (although without any trace whatsoever 

 of a fascicular composition) is represented, or the round form is 

 selected without definite intention, is, I think, very doubtful. 



The height of the shaft, referred to that of the column, is surprisingly con- 

 stant, varying between .72 and .77, — which approaches the corresponding pro- 

 portions of the lotus columns of Order 1. 



The shaft-height, measured by its greatest diameter, falls in some rather 

 doubtful examples to 3.52 and 3.65, but varies in better authenticated ones be- 

 tween 4.15 and 5.50. These figures will be seen to range much higher than the 

 corresponding ones for the papyrus group of Order I., and to approach those of 

 the lotus group. 



The diminution of the shaft is very irregular. In general it is much less 

 than before, and in later examples the shaft verges upon cyliudricity. 



Astragal. — The astragal here is nearly the same as in the group 

 called B. 2 in Order I. It is either very superficially represented or 

 altogether omitted. Its position on the shaft is not quite constant ; in 

 later forms it is often separated from the capital by an interval about 

 ecjualliug its own width.* 



The width of the astragal falls between .075 and .111 of the shaft-height. 



There are no proper astragal pieces. 



As was just remarked, there is often an interval between the 

 astragal and the capital. In this interval there seems to be a reversion 

 to the bundle idea, for it is usually occupied by vertical stalks, sharp- 

 edged or rounded, but nearly alw T ays small and numerous.! Yet the 

 reversion is more apparent than real, for from the facts that bundle 

 columns are not produced at the same period, that the most typical of 

 the early forms lack these stems, that the stems themselves are so 

 small, and, finally, that their position is ordinarily determined by that 

 of the ornaments of the capital, I am convinced that they must be 

 regarded as appendages of the capital, bound upon the outside of the 

 shaft by the astragal. But although essentially a part of the capital, 

 since its practical effect is to prolong the shaft beyond the astragal, I 



* Description, i. 8, figs. 6, 7, 9-15. 



t Ibid., i. 8, fig. 12. In one instance a zone of triangular scales, similar to 

 those presently to be described as belonging to the Palm. Order, takes the place 

 of these stems. See Ibid., i. 75, fig. 2. 



