that there had been a publication of the work. The poet's scheme may be discerned in the account he gave to Allen. He informed him that the book, being most of it printed, was "put past preventing," but that he was "trying all the means possible to retard it." In plain words, he was manoeuvring to keep back the Irish edition till his rival reprint was in the market. When he had succeeded in his device, he repeated his old tactics of advertising that the surreptitious collection was the cause of his own, and at the same time bespoke the preference for his reprint by announcing that it would contain "several additional letters."[150] Apart from these additions, the quarto of Pope is a reproduction, with some variations, of the Dublin impression, and a few notes which Faulkner had doubtless found in the volume sent from England, are said in the quarto to be taken from Faulkner. Nevertheless there is strong internal evidence that a portion of the quarto had an independent origin, and had been printed off before the Irish edition was received. The correspondence consists of 209 pages, which are numbered consecutively from 1 to 115. At this point the letters of Swift to Gay commence, and instead of the numbers proceeding in regular order, they go back to page 89, and are thence continued without any break to the final page, 182. That the arrangement is not a typographical mistake is clear from the signatures of the sheets being in accordance with the paging,--a coincidence which was barely possible if the figures had been a misprint. The correspondence of Swift with Gay begins on sheet N, which is the letter of the alphabet that answers to page 89 in a quarto volume, and this keeping between the letters and the figures is preserved throughout. But there is a second coincidence which is absolutely fatal to the idea that the confusion in the paging was an error of the press. The quarto edition was accompanied by an edition in folio, which was the same impression with the matter parcelled out into pages of greater length, and with the requisite changes in the numbering of the pages and the signatures of the sheets. In spite of the change there is the identical peculiarity that distinguishes the quarto. The numbers run on unbroken from 1 to 108, when we arrive at the letters to Gay. Here we recommence with page 85, and starting from this new basis the figures proceed in regular succession to the end. The sheet at page 85 is marked Y, the proper letter for the folio size, and as in the quarto the signatures, in every instance, correspond with the pages. The defect cannot be explained by the supposition that the work had been divided into portions, which were printed separately for the sake of expedition. With the text of the Dublin copy to guide his calculations, no compositor could have committed the error of pronouncing that matter which covers 115 pages could he contained in 88. The evident cause of