supplied, along any given line. We will assume that either in the ways indicated, or in some other, the librarian has satisfied himself that he understands what his community needs. How shall he find the books that will satisfy that need, and when they are found (or, still more, when they obtrude themselves on his notice) how shall he know that they are what they claim to be? In order to find what he wants, the librarian naturally turns at first to such classed bibliographies as he has at hand, including publishers’ trade lists. Unfortunately, books very rapidly become out of print, and if his bibliography or list is even two or three years old he cannot be sure that his work of selection is not in vain. The value of the A.L.A. catalog has been much impaired by its inclusion of out-of-print books, and as, now that it is several years old, the number of these is increasing daily, its use has become more and more vexatious, both to librarians and publishers. It is to be hoped that in the new edition now preparing the out-of-print books will be omitted. Fortunately we now have at our disposal yearly alphabetical lists of in-print books. Such are the index to the Trade list annual and the United States catalog for American editions, and the Index to the reference catalog of current literature for British books. If the needs of your library require that some one class should be largely replenished, you may call in expert knowledge. Some teacher or student who is a specialist in that subject is generally not hard to find, and his advice will be of the greatest value. Special bibliographies are valuable in inverse ratio to their length--a complete list of works on Egyptology, for instance, is hardly more valuable to the ordinary small library than a full, unclassified list of books in-print on all subjects. The majority of the small library’s purchases are books as currently issued. For these the _Publishers’ weekly_ is indispensable. Some librarians prefer to look at every book before purchasing, and arrange with publishers or booksellers to send large numbers of books weekly or even daily on approval. This, if there is sufficient time, is a good plan, but it is certainly wasteful. There are many books which we can surely reject or accept from the author and title entry in the _Publishers’ weekly_ as well as if the actual book were in hand. If a mistake is made it will be, or should be, discovered as soon as the book is received, and the volume can then be exchanged. Only the doubtful books need be asked for on approval, and these will generally be found to constitute a relatively small percentage of the whole.