ORNAMENTATION OF CARPETS COLE. 127 



are almost unintelligible, though the original of them probably was 

 a dragon's head; the dragon was invented by the Chinese almost as 

 early as the Sphinx was invented by the Egyptians, and apparently 

 some centuries before Perseus encountered any similar creature. 



The next slide shows a simple but adequate frame of the sort 

 which has been in use from old times by wandering families or 

 groups of carpet makers in Turkestan, farther east, and south. 

 In such a frame flat or raised surface rugs could be made. These 

 wandering weavers have inherited, as it were, the designs they work 

 in their rugs; and, unless they come into the service of some merchant 

 or patron who furnishes them with other designs, they continue to 

 produce with scarcely any intended, but with a good deal of acci- 

 dental, variation of their own traditional patterns and designs. And 

 this condition has lasted amongst such peoples for many centuries. 



This slide is from the carving of a floor covering which was prob- 

 ably of tapestry weaving, as indeed was the greater number of orna- 

 mented textiles made by Egyptians, Assyrians, Persians, and Greeks 

 before the Christian era. This carving was discovered in the ruins 

 of Kouyunjik and is of Assyrian workmanship, eighth century B. C. 

 The plan of its design, as fully displayed in the whole of the floor 

 covering, originally corresponded with that of the Kurdish rug, hav- 

 ing its field of pattern inclosed within a border. In this case the 

 ornamental features of the border are well shaped, and are based 

 upon plant forms. The outer ones are alternately buds and ex- 

 panded flowers, those in the next series are full daisy blossoms, and 

 then come repeated palm or radiating palmette forms. The pattern 

 of the field is formed with intersecting circles, and is a truly abstract 

 pattern, being unrepresentative of any actual things and not sym- 

 bolical in any waj\ The texture of such a carpet was, as I have 

 said, probably that of tapestry weaving and not of raised or cut pile. 

 Indeed, the manufacture of this latter and more complicated material 

 does not seem to have been known by the old Egyptian, Assyrian, 

 Persian, and Greek weavers. The nearest approach to raised surface 

 textiles made by them were linen cloths faced with loose loops. 

 These give a shaggy-faced material resembling modern bath towels. 

 Several pieces of it have been found in disused Egyptian cemeteries, 

 dating probably from the first century B. C. or A. D., and. it is con- 

 sidered by various authorities that they are identical with a fabric 

 called by Aristophanes " Persis," and reputed as a manufacture of 

 barbarians. The Greeks, however, also manufactured similar tex- 

 tures, and called them " kaunakes " and " phlocata." Pliny, writing 

 500 years later, mentions corresponding stuff as " amphimalla " when 

 the shagginess was on both of its sides, and " gausapa " when woven 

 on one side only. This shaggy material was apparently as common 

 in use as tapestry weavings, but it does not seem to have lent itself 



