We must, however, bear in mind that the present text of the LXX. is of all texts the most time-worn, and often requires correction. Most providentially we do not depend on one recension; we have in existing copies the remains of several, and we may make use of them to restore the original readings. The problem in this case differs essentially from that which meets us in revising the Greek text of the New Testament. Here diplomatic evidence has not the same weight as there. Emendations may be detected, even when better readings of the Hebrew, by want of conformity to the Hebrew order (and the temptation to make such kind of alterations, when the version was used independently, would clearly be very great), or again a comparison of the various readings will enable us to guess with tolerable confidence what the Hebrew originally was. In this way, when the Hebrew text is doubtful, we can ascertain the correct reading by searching for that common origin from which the variants in the LXX. were derived, and we may then turn round on the version with the help of the Hebrew, and show how the changes successively arose. An instance of this will be found at chapter x. 10; how far I have succeeded the reader must judge. We must also bear in mind that what we should now denote by marginal renderings are in the ancient versions inserted in the text. The interlinear character of the version enables us to detect this: we find two Greek words standing in place of a single Hebrew equivalent. The result of all this, as applied in the Book of Ecclesiastes, is to vindicate the accuracy of the received Hebrew text, and, in a less degree, of the pointing and accentuation also. Only in a very few instances is it needful to propose an emendation of the Hebrew text, and that where the ancient versions are apparently unanimous in requiring it.