So far as the subject matter of the book is concerned, my test would be simply that of its effect on the reader. If a book makes the reader want to be mischievous, foolish or criminal--to be a silly or bad man or woman, or if it tends to make him do his daily work badly, it is a bad book and all the worse in this case if it is interesting and fascinating in style. But even here the trouble is largely in the manner of treatment. A book may tell of crime and criminals in such a way as to make the reader detest both or feel an attraction toward both. In this case, as the scripture says, “Ye shall know them by their fruits.” If a book sends a boy out to be a burglar, it is bad; if it impels him to take a crying child by the hand and lead it home, it is good. And here let me say that this compelling power, this effective result of a book should speak in its favor though all other tests be against it. Musicians tell us that a great composer may write a work that breaks every rule of harmony and yet be a work of genius. Genius knows no rules. So much for the general line of cleavage. But the special may for the moment exclude all the claims of the general. A community may be in crying need of books on a given subject--pottery or rowboats or hygiene. This need may or may not be realized by the community, but its existence makes a special class of books the best, for the moment, for that community. To buy a good collection of minor poets for a town that clamors, or ought to clamor, for books on the electric industries, is to get bad books. Now do we, under our present system, or lack of system, in selection, get these best books--best both in the general and in the special sense? What is the matter with the books in the average small library? The trouble is not generally that the books are bad, but that they might easily be better, and by “better” it must be borne in mind that I mean more closely adapted to the legitimate needs of the community. If we go over the shelves of the average small library we shall generally be able to note the following facts: (1) A considerable portion of the books have not been taken out in long periods. This can easily be ascertained by examining the book-cards or dating-slips. Of course, the non-use of a book does not mean that it should not be in the library. The fault may be with the readers, not with the book. Non-use, however, does mean that something is the matter. Either the library public has had taste or is not properly guided, or else a mistake was made in providing it with this particular book. (2) A considerable number of standard books whose reading should be encouraged will not be found on the shelves. These books are almost always part of the collection, but there are not enough duplicates to supply the demand. At the same time it will be found that the library is adding current books of doubtful value.