to obtain as soon as he can, by the severest economy, a restricted and steadily increasing series of books, for use through life; making his little library, of all his furniture, the most studied and decorative piece." And Henry Ward Beecher urged it as the most important early ambition for clerks, working men and women, and all who are struggling up in life, to form gradually a library of good books. "It is a man's duty," says he, "to have books. A library is not a luxury, but one of the necessaries of life."

And says Bishop Hurst, urging the vital importance of wise selection in choosing our reading: "If two-thirds of the shelves of the typical domestic library were emptied of their burden, and choice books put in their stead, there would be reformation in intelligence and thought throughout the civilized world."

SELECTION OF BOOKS FOR PUBLIC LIBRARIES.

Let us now consider the subject of books fitted for public libraries. At the outset, it is most important that each selection should be made on a well considered plan. No hap-hazard, or fitfully, or hastily made collection can answer the two ends constantly to be aimed at—namely, first, to select the best and most useful books, and, secondly, to economize the funds of the library. No money should be wasted upon whims and experiments, but every dollar should be devoted to the acquisition of improving books.

As to the principles that should govern and the limitations to be laid down, these will depend much upon the scope of the library, and the amount of its funds. No library of the limited and moderate class commonly found in our public town libraries can afford to aim at the universal range of a national library, nor even at the broad selections proper to a liberally endowed city library.

But its aims, while modest, should be comprehensive enough to provide a complete selection of what may be termed standard literature, for the reading public. If the funds are inadequate to do this in the beginning, it should be kept constantly in view, as the months and years go on. Every great and notable book should be in the library sooner or later, and if possible at its foundation. Thus will its utility and attractiveness both be well secured.

Taking first the case of a small public library about to be started, let us see in a few leading outlines what it will need.

1. A selection of the best works of reference should be the corner-stone of every library collection. In choosing these, regard must be had to secure the latest as well as the best. Never buy the first edition of Soule's Synonymes because it is cheap, but insist upon the revised and enlarged edition of 1892. Never acquire an antiquated Lempriere's or Anthon's Classical Dictionary, because some venerable library director, who used it in his boyhood, suggests it, when you can get Professor H. T. Peck's "Dictionary of Classical Antiquities," published in 1897. Never be tempted to buy an old edition of an encyclopaedia at half or quarter price, for it will be sure to lack the populations of the last census, besides being a quarter of a century or more in arrears in its other information. When consulting sale catalogues to select reference books, look closely at the dates of publication, and make sure by your American or English catalogues that no later edition has appeared. It goes without saying that you will have these essential bibliographies, as well as Lowndes' Manual of English Literature first of all, whether you are able to buy Watt and Brunet or not.

2. Without here stopping to treat of books of reference in detail, which will appear in another place, let me refer

to some other great classes of literature in which every library should be strong. History stands fairly at the head, and while a newly established library cannot hope to possess at once all the noted writers, it should begin by securing a fine selection, embracing general history, ancient and modern, and the history of each country, at least of the important nations. For compendious short histories, the "Story of the Nations" series, by various writers, should be secured, and the more extensive works of Gibbon, Grote, Mommsen, Duruy, Fyffe, Green, Macaulay, Froude, McCarthy, Carlyle, Thiers, Bancroft, Motley, Prescott, Fiske, Schouler, McMaster, Buckle, Guizot, etc., should be acquired. The copious lists of historical works appended to Larned's "History for Ready Reference" will be useful here.

3. Biography stands close to history in interest and importance. For general reference, or the biography of all nations, Lippincott's Universal Pronouncing Dictionary of Biography is essential, as well as Appleton's Cyclopaedia of American Biography, for our own country. For Great Britain, the "Dictionary of National Biography" is a mine of information, and should be added if funds are sufficient. Certain sets of collective biographies which are important are American Statesmen, 26 vols., Englishmen of Letters, - vols., Autobiography, 33 vols., Famous Women series, 21 vols., Heroes of the Nation series, 24 vols., American Pioneers and Patriots, 12 vols., and Plutarch's Lives. Then of indispensable single biographies there are Boswell's Johnson, Lockhart's Scott, Froude's Carlyle, Trevelyan's Macaulay, Froude's Caesar, Lewes' Goethe, etc.

4. Of notable essays, a high class of literature in which there are many names, may be named Addison, Montaigne, Bacon, Goldsmith, Emerson, Lamb, De Quincey, Holmes, Lowell, etc.

5. Poetry stands at the head of all the literature of imagination. Some people of highly utilitarian views decry poetry, and desire to feed all readers upon facts. But that this is a great mistake will be apparent when we consider that the highest expressions of moral and intellectual truth and the most finely wrought examples of literature in every nation are in poetic form. Take out of the world's literature the works of its great poets, and you would leave it poor indeed. Poetry is the only great source for the nurture of imagination, and without imagination man is a poor creature. I read the other day a dictum of a certain writer, alleging that Dickens's Christmas Carol is far more effective as a piece of writing than Milton's noble ode "On the Morning of Christ's Nativity." Such comparisons are of small value. In point of fact, no library can spare either of them. I need not repeat the familiar names of the great poets; they are found in all styles of production, and some of the best are among the least expensive.

6. Travels and voyages form a very entertaining as well as highly instructive part of a library. A good selection of the more notable will prove a valuable resource to readers of nearly every age.

7. The wide field of science should be carefully gleaned for a good range of approved text-books in each department. So progressive is the modern world that the latest books are apt to be the best in each science, something which is by no means true in literature.

8. In law, medicine, theology, political science, sociology, economics, art, architecture, music, eloquence, and language, the library should be provided with the leading modern works.

9. We come now to fiction, which the experience of all libraries shows is the favorite pabulum of about three readers out of four. The great demand for this class of read

ing renders it all the more important to make a wise and improving selection of that which forms the minds of multitudes, and especially of the young. This selection presents to every librarian and library director or trustee some perplexing problems. To buy indiscriminately the new novels of the day, good, bad, and indifferent (the last named greatly predominating) would be a very poor discharge of the duty devolving upon those who are the responsible choosers of the reading of any community. Conceding, as we must, the vast influence and untold value of fiction as a vehicle of entertainment and instruction, the question arises-where can the line be drawn between the good and improving novels, and novels which are neither good nor improving? This involves something more than the moral tone and influence of the fictions: it involves their merits and demerits as literature also. I hold it to be the bounden duty of those who select the reading of a community to maintain a standard of good taste, as well as of good morals. They have no business to fill the library with wretched models of writing, when there are thousand of good models ready, in numbers far greater than they have money to purchase. Weak and flabby and silly books tend to make weak and flabby and silly brains. Why should library guides put in circulation such stuff as the dime novels, or "Old Sleuth" stories, or the slip-slop novels of "The Duchess," when the great masters of romantic fiction have endowed us with so many books replete with intellectual and moral power? To furnish immature minds with the miserable trash which does not deserve the name of literature, is as blameworthy as to put before them books full of feverish excitement, or stories of successful crime.

We are told, indeed (and some librarians even have said it) that for unformed readers to read a bad book is better