IV 



Now these cyclopedias , in their turu, have been composed in nearly the sarae 

 way. Professing to embrace a certain range of subjects, some of which are treated 

 with considerable skill, the notices about foreign countries are introduced chiefiy for 

 the sake of completeness , and as a rule the author does not seem to have bestowed 

 much research upon them. He just takes the Dynastie Histories and composes his 

 account from the notices he finds there, abbreviating more or less , and frequently 

 committing serions errors by fanciful corrections of what he does not understand ; he 

 throws together and mixes up the information of different times , which thereby loses 

 its fixed dates and with them much of its value. We have consulted the principal of 

 these cyclopedias , without fmding material for a single extract. 



Under these circumstances it need not astonish us, that the accounts translated by 

 Amiot and Schlegel were very defective and unsatisfactory , and that their translations 

 are disfigured by grave errors , some of which we shall have to correct in the following 

 pages. We feel obliged to say however that many of these errors are less the fault of 

 the trans lators , than of the worthless material on which they bestowed their pains. 



Leon de Rosny has made his translations from the Cyclopedia San-ts'ai T'u Hui, 

 which is amongst the least complete on this subject, and moreover he has committed 

 many mistakes of his own : he places Pekalongan on Borneo and Tuban on Sumatra , 

 Chan-ch 1 éng and Champa are made two different places , he Avrites Kuawa instead of 

 Java , Tu-po instead of Dja-po or Dja-pa, etc. His mistakes have been caused by in- 

 sufficiënt acquaintance with the geography of these countries , whilst a part of them 

 might have been avoided , if he had compared other sources with the one he has translated. 



Our plan has been different from that of our predecessors. We have made a 

 collection, as complete as we coulcl, of the literature on the subject, and by reading 

 everything, we have generally been able to tracé the different notices to their first 

 appearance and to ascertain with more or less certitude the time to which their refer. 

 By this process of comparison it was also possible to estimate the accuracy and value 

 of every account , whilst many passages , unintelligible at first sight, became clear in 

 the course of our reading. We translated only what was original , and the endless re- 

 petitions , through which we were obliged to go, were consigned to oblivion. In this 

 way it has been necessary to read at least ten times more than what has been trans- 

 lated ; our task has been long and often became tedious , but we see no other course 

 to a reliable result. 



Whilst thus recommending our method, we hope it will not be defmitely 

 judged by the results we have obtained with it. The Malay countries have been 

 treated -first for special reasons , which it is not necessary to explain here , but being 

 situated at a considerable distance from China, they were very imperfectly known 

 there , and the way in which they are noticed in the geographical literature of the 

 Chinese leaves much to desire. As soon however as a country nearer to China is 

 taken ,' there is a considerable improvement , and for the different countries of Indo- 

 China we have found the sources of information much more abundant and complete. 



