Much effort is being expended by libraries at the present time in promoting the reading of their books. It is being more and more recognized that a smaller number of books more widely read fulfills the real purpose for which the library exists better than a larger number standing on the shelves. This is now so much of a commonplace that we are liable to forget how new the idea is. It was not so long ago that the annual report pointed with pride to the large proportion of income spent on books and the small amount on administration. The whole movement expressed by the term, "publicity," is the growth of a few years. So far most of our work along this line has been devoted to promoting the reading of new books and technical works. Gratifying success has crowned our various schemes. But every library worker knows that the easiest class of books for which to find readers is new books. The reasons for this are so apparent that we need not dwell upon them. To circulate the great books, the classics, the books which constitute literature in the restricted sense is another and a far more difficult undertaking, and on this we have hardly made a beginning. Yet if the library is to stand--and we all believe it should--for the highest, for true culture and refinement, if it is to be a source of ideals, as well as ideas, here is a side of our work which must not be neglected. We may be inclined at times to underrate the library's ability to secure the reading of specific books. An experiment tried some years ago may serve as an object lesson. Van Vorst's "The woman who toils" and "The souls of black folk," by Du Bois, were selected for this experiment. Under ordinary conditions the first of these books would have enjoyed a fair degree of popularity, while the second would have had a rather small circulation. The library bought a number of copies of each, sent notices to all the papers, had book-notes in its bulletin, put up publishers' advertisements on its bulletin-boards, and (note this last) discussed the books in staff-meeting so that every assistant was able to talk about them intelligently. The results surpassed expectations. For months it was impossible to meet the calls for them, and reserves came in steadily; most remarkable of all, after eight years the circulation of one is eight and the other three times above the average. So much for what a library can do in determining what its constituency shall read.