192 ON THE ANCIENT GEOGRAPHY 



the known world, should appear on the map as an entire blank. 

 As they were far, however, from being all perfectly known, it 

 became often necessary, in acting upon this principle, to have 

 recourse to somewhat arbitrary measures. One of the most 

 common was, by extension or repetition, to fill up the un- 

 known parts out of the known, a process of which many in- 

 stances must have occurred, to those who are versant in such 

 researches. The combined celebrity and obscurity of the pre- 

 sent object, might seem to authorise more than the usual li- 

 cence* Accordingly, it appears to me, that Ptolemy has en- 

 tirely protracted, into Serica, the geographical lines of Eastern 

 Scvthia. Of this operation there appear, in the very descrip- 

 tion, to be pretty evident traces. We have Issedon Scythica, 

 Issedon Serica ; while, of the Casian, Auzacian, and Emodian 

 mountains and regions, the western part is in Scythia, and the 

 eastern in Serica. It is, I think, quite unexampled, that any 

 great country, much less so immense a country as Serica, and 

 one separated by such powerful barriers from the rest of the 

 world, should consist almost entirely of districts protracted out 

 of another region, with which it has no natural or political con- 

 nexion. We may observe, that Ptolemy stands alone in re- 

 presenting Serica to be a country of mountains. Ammianus, 

 as a historian, has given, at some length, a description of its 

 general aspect. His representation decidedly is, that of a plain 

 of vast extent and luxuriant fertility, and which was only sur- 

 rounded by a circuit of mountains. I think it even appears, 

 that the same idea was in Ptolemy's mind; for he begins by 

 saying, " Mountains surround the Seres ;" though in the deli- 

 neation he certainly never reaches beyond these mountains. 



From the whole of what has now been said, it may be infer- 

 red, that all the details given respecting Serica are to be sought, 

 and may perhaps be hereafter found, with some degree of pre- 

 cision, 





