as he left it at his death in 1894, consisted of about 2500 folios. Had Whitney lived to see it printed, the editor of this Series would probably have read one set of proofs, and made suggestions and criticisms freely on the margins, which the author would then have accepted or rejected without discussion; and the whole matter, in that case a very simple one, would have been closed by a few lines of kindly acknowledgment from the author in his preface.
Picking up the broken threads.—It is, on the other hand, no simple matter, but rather one of peculiar difficulty and delicacy, to edit such a technical work as this for an author who has passed away, especially if he has been the editor's teacher and friend. The difficulty is increased by the fact that, in the great mass of technical details, there are very many which have to be learned anew by the editor for himself, and others still, which, through long years of labor, have grown so familiar to the author that he has hardly felt any need of making written memoranda of them, and which the editor has to find out as best he can.
Relation of the editor's work to that of the author.—Although Whitney's manuscript of the main body of the work was written out to the end, it was not systematically complete. Thus he had written for book i. (and for that only) a special introduction, showing that he meant to do the like for the other eighteen. Of the General Introduction as it stands, only a very few parts were worked out; for some parts there were only rough sketches; and for very many not even that. And in unnumbered details, major and minor, there was opportunity for long and patient toil upon the task of systematically verifying all references and statements, of revising where need was, and of bringing the whole nearer to an ideal and unattainable completeness. What these details were, the work itself may show. But besides all this, there was the task of carrying through the press a work the scientific importance of which called for the best typographical form and for the utmost feasible accuracy in printing.
Parts for which the author is not responsible.—No two men are alike in the various endowments and attainments that make the scholar; and, in particular, the mental attitude of any two towards any given problem is wont to differ. It is accordingly not possible that there should not be, among the editorial additions to Whitney's manuscript or changes therein, many things which he would decidedly have disapproved. They ought certainly therefore to be marked in such a way that the reader may easily recognize them as additions for which the editor and not the author is responsible; and for this purpose two signs have been chosen, ⌊ and ⌋, which are like incomplete brackets or brackets without the upper horizontal strokes, and which may be called "ell-brackets" and suggest the
