Hsiang that have come down to us. It is impossible to say, from the want of information, what liberties he took with the documents put into his charge. The differences between the two texts as we now have them are trivial. I believe that the changes made in them by Hsiang were not important; but having them as they came from his revision, we have them at second hand, and this has afforded ground for the dealing with them by Kû Hsî and others in the manner which will be described in the next chapter.
2. I have said above (p. 450) that for the text of the classic,—the modern text, that is,—as we now have it, we are indebted to the labours of the emperor Hsüan Zung of the Thang dynasty. Kû Î-tsun, of the Khien-lung period (1736–1795), in his work on the classics and the writings on them, has adduced the titles of eighty-six From Khung An-kwo to the emperor Hsüan Zung. different works on our classic, that appeared between Khung An-kwo and Hsüan Zung. Not a single one of all these now survives; but the enumeration of them shows that the most distinguished scholars during the intervening centuries exercised their powers on the treatise, and would keep a watch on one another in the preservation of the text. Moreover, several of the works continued through the Thang dynasty, and on into that of Sung. The Catalogue of the Sui Library contains the titles of nineteen in its list.
The emperor Hsüan says, in his preface, that in the making of his commentary he had freely used Hsüan Zung's work. the commentaries of six earlier writers, whom he names. They were, Wei Kâo, Wang Sû, Yü Fan, and Liû Shâo, all of our second and third centuries; Liû Hsüan, of our sixth century, who laboured on the commentary of Khung An-kwo, which, as I have already stated, is said to have been discovered in his time and presented to him; and Lû Khang, rather earlier than Liû, who dealt critically with the commentary attributed to Kăng Khang-khăng. 'But,' says the imperial author, 'if a comment be right in reason, why need we enquire from whom it came? We have therefore taken those six writers, considered wherein