334 REPORT OF NATIONAL MUSEUM, 1900. 



or Persian records of that day may yet supply the missing explanation. 

 Whatever it was, the supply was exhausted during the Ch'enghua 

 period. Somewhat later, however (during the Chengte period, 1606 to 

 1521), Tatang, the governor of Yunnan province, succeeded in obtain- 

 ing further supplies of Mohammedan blue b}^ paying for it twice its 

 weight in gold; and during this and the greater part of the subsequent 

 reign (Chiuching period, 1522 to 156(5) it continued available— a fact to 

 which is doubtless attributable the excellent color of the productions 

 of that time. Tow^ards the close of the latter reign, however, the sup- 

 pi}' again gave out, when an incinerated cobaltiferous ore of manga- 

 nese (termed wu-ming-t) replaced the western product; the color 

 obtained from this native ore, far from equaling the brightness and 

 transparency of the foreign blue, however, showed a dull and heavy tint 

 after baking. 



EGGSHELL PORCELAIN. 



Eggshell porcelain of ver}' delicate workmanship was produced, but 

 owing to its extreme fragility good specimens are now difficult to 

 obtain. It appears also to have had a tendency to crack during the 

 process of firing. These porcelains are termed among the Chinese 

 t^'o-fai^ or porcelain from which the "embr}^©" or biscuit has been 

 removed, and are divided into two classes: "True ^'(9-^'«^,'' ^^ very 

 thin, also known as eggshell {tan-p''i or luan-imi)^ and "semi t^o-f'ai^'' 

 the somewhat thicker. The true f'o-f'ai especially present great diffi- 

 culties in the manufacture and require extraordinary dexterity in the 

 handling, for so thin is the portion of the body the workman allows to 

 remain that it seems as though all had been removed; and it is only 

 quite recently that the Government manufactory at Sevres has succeeded 

 in producing such porcelain, and then by an entirely diiferent process — 

 by casting or moulage en harbotine. 



The work translated by M. Julien states that while the production of 

 this ware originated during the Yunglo period, it was only the thicker 

 variet}^ that was then made, and that the true t''o-t^al dates from a later 

 epoch, having been produced during the Ch'enghua period (1465 to 1187) 

 at the government manufactory and during the Lungch'ing (1567 to 1572) 

 and Wanli (1573 to 1619) periods at private factories. This statement 

 appears, however, to be erroneous; for in No. 295 of this collection will 

 be found a specimen, so at least Chinese experts state, of the serai t''o-t''ai 

 (though it seems difficult to believe that a bowl of such size could be 

 made nmch thinner and yet be of practical utility), and in Nos. 289 to 

 291 specimens of the true t''o-t''ai^ both having the inscription Yung-lo- 

 nien-c?dh, "Made during the Yunglo period," engraved in the old seal 

 character on its foot. Moreover, the one specimen of this ware described 

 by Hsiang Tzu-chiug is a small cup "as thin as paper," called t''o-t''ai^ 

 "bodyless,"^. 6., true f" o-t''ai, not SQmi f'o-t^ai^ of which he saj's "there 



