Page:History of Art in Persia.djvu/87

 Construction. 73 It is a style frequently introduced in fortification walls by modern engineers. No inference. therefore> is to be drawn from the fact that the courses are more or less regular, or of discrepancies which in this instance are of no moment The Persian builder, like his modern ton/rere, employed blocks of varying shape and size, as best suited his purpose. Compare, for example, the brace of funereal towers at Parsagadae and Naksh-i-Rustem (Fig. 21). They are built on the same plan, and, despite slight irregularities, both evince a marked tendency to horizontal courses. There is a curious constructive detail about these towers which has not yet been satisfactorily explained. On the four sides of the wall appear rectangular incisions, whose sunken faces, it has been urged, were to act as landmarks for cutting away the stones surrounding them to an even surface.* What tells against the conjecture, is the fact that the wall surface has all the appearance of having been smoothed over and dressed with the same amount of care as the supporting pilasters at the sides, about which no such depressions occur. Besides, is it conceivable that if they were not destined to last, but would naturally disappear as soon as the dressing of the stones was finished,* the builder would have taken so much super- fluous trouble in cutting them to a uniform size and shape. The saliences occasionally encountered in unfinished Greek work are far removed from such a regularity as this. Again, it would be strange, to say the least of it, that in both towers the masons should have stopped at precisely the same point We incline to ascribe a decorative function to the incisions under notice, made for the sake of breaking the monotony of a large plain surface.' The question, too. may be asked, whether these hollows were not fitted with some material other than stone ; such as coloured or enamelled slabs, or perhaps black marble. Our hypothesis would account both for the great number of these hollows, the uniformity of their size and symmetrical dis- tribution. A thorough search among these ruins and their surroundings might, perhaps, bring to light fragments of a decoration which we think existed here. Another problem, of far greater import, is one which every ■ DiEULAFOY, L'Ari antique, etc., torn. i. p. 16. ' 3id. IKeuUfoy does indeed assign to tbem a decorative character, but in his opinion it was sobordinate to the purpose for which they were made. Digitized by Google