would hardly demand discussion, but for the possibility that it may after all be forced upon the Museum, notwithstanding its repugnance to the common-sense of the late and the present Principal Librarian. If ridicule could kill, it could hardly have survived the discussion which arose among its advocates at the late Oxford Conference. Those external to the Museum suggested that the Museum should catalogue not only the old English books it possessed, but also those it did not possess. The Museum representatives, enamoured with the project as they were, pleaded that it would be unreasonable to expect them to describe what they had never seen. The other side concurred, but represented in turn that a catalogue of such English books only as happened to be in a particular library would be very imperfect, and of very little use. Having thus mutually demonstrated the unreasonableness of the proposal from one point of view, and its inutility from another, they agreed that it should by all means be persevered with, and went home. The subject of the classification of books within the library itself—a matter of even more importance to the librarian than the preparation of classified lists—has received a great impulse from the ingenious system contrived by the principal editor of the _Library Journal_, Mr. Melvil Dewey. Mr. Dewey—a remarkable instance of the combination of disinterested enthusiasm with thorough business capacity—is devoted to several other causes beside the causes of libraries, and among these is the cause of the decimal system. His experience in the latter field has given him the idea of dividing the departments of human knowledge decimally. His scheme provides for a thousand divisions. Every tenth number embraces some important section of knowledge, and the following nine as many subjections or allied subjects admitting of classification under the principal head. Thus number 500 might represent mathematics in general, and 501 conic sections, analytical geometry, or any other branch of the general subject. Further subdivisions, if needed, would be made by appending letters to these numerals, as 501_a_, 501_b_. Each book would be numbered in the order of its accession to the library, and receive its place upon the shelves accordingly, so that there never would be any doubt as to the press-mark or position of a book that had once been properly classed. Our space does not permit us to dwell upon many other points connected with the working of this ingenious scheme, which, if inapplicable to the great old libraries whose catalogues, like the Abbé Vertot's siege, are already done, deserves the most careful consideration on the part of the founders of new institutions. It must, as the inventor admits, receive some modification in practice from the impossibility of accommodating books of all sizes upon the same shelf; it is only to be feared that these and similar necessary condescensions