6 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 79 



waringau tree, possibly a ficus. It appears from the specimens of 

 cloth from different localities that several species of trees furnish 

 bark, as some of the cloth is coarse and some quite fine; none, how- 

 ever, having the softness of Samoan or especially Fijian tapa. The 

 refinements observed in Polynesian bark cloth are not found in the 

 more primitive examples of Malaysia. The Malays have not dis- 

 covered or jDerhaps needed the method of increasing the size of the 

 cloth by pasting strips together as in Polynesia, where sometimes 

 immense pieces are made in this way. The Malay economy did not 

 require spreads or partitions or costumes of this material. Curi- 

 ously, in the Congo tapa is joined by sewing together pieces of 

 regular size as in our familiar patchwork quilts. 



Some patterned tapa beaters are seen in the Celebes, but not of the 

 fineness of those found in Polynesia, where the refinement of club 

 pattern marks in the texture of the cloth is carried to a high point. 

 The pattern marks resemble watermarks in paper and are likewise 

 generally seen only by holding the fabric up to the light. Another 

 refinement in Polynesia is seen in the perfuming of tapa cloths. 



No decorating devices are seen in Malaysia as the Samoan printing 

 blocks or the Fijian carved bamboo strips. The decoration of Celebes 

 and other JNIalay tapas is simply done free-hand with sticks dipped 

 in color. No continuous border lines appear. In most cases the 

 designs are painted on the tapa as individual units, that is, sprigged 

 as in Hindu muslins, and with no intent to produce an all-over deco- 

 ration as in most Polynesian tapas. 



The Celebes decorated bark cloth shows a limited range of color, 

 soft red, yellow brown, and black being used. These are evidently 

 vegetal colors derived from the plant environment. Aniline colors 

 have found their way here and especially at Bada Toare are the for- 

 eign dyes seen. As a rule the colors are semiskillfully laid on. In 

 some instances the drawing is rather good. In general, while the 

 water-buffalo art has found itself in design, the execution has lagged. 

 It does not compare with the sure and intricate art of Borneo, and 

 this is a point bearing on whether the Celebes art discussed here 

 shows traces of a comparatively recent origin. It has been stated 

 that a large number of the specimens of bark cloth collected by Mr. 

 Raven are for winding about the head ; that is, turbans. This form 

 of headdress which, so far as the writer is informed, has not been 

 studied, is undoubtedly ancient. Its extension from its presumed 

 points of origin in the Near East appears to be due to the INIoluun- 

 medan concjuest. The causation origin of the turban would be pro- 

 tection from a fierce sun. It is possible, then, that the Malay turban 

 would come into use in this remote reo-ion some time after the ninth 



