gestions; the very fact that it makes no pretensions to style enables its maker to crowd information into it without occupying an amount of room that would alarm the economical. In what may be called the technique of cataloging there are several doubtful questions, and I have my doubts whether common sense has always been sufficiently consulted in the formation of the generally accepted rules; or perhaps I should say, whether rules adopted by the highest authority for a library of the size and character of the British Museum are adapted to the needs of every town library. I can only glance at a few points. the majority such things are as if they were In classification, too, common sense has many a word to say. It does not strike me as a very sensible proceeding to classify books on the shelves systematically, and then to classify them in the catalog on the same system, making the catalog only a glorified shelf-list, when one might get another kind of information by arranging the catalog differently, in alphabetical subject order, for instance. It does not seem to me sensible to divide a very small library into very numerous subdivisions, requiring very long class marks, or to use in any library, large or small, divisions which are not so well defined as to be easily applied by the classifier and easily understood and used by the public. Still less does it seem reasonable to me to divide a growing library as if it were always to remain of its present size, and to apply to it a system which is incapable of expansion and further subdivision. Cataloging-rule makers have always shown a tendency to seek what was easiest for the cataloger, as if it were for him that the catalog is made, and not for the public. If the object of the catalog is to enable the public to find books quickly, surely entries should be made under the name by which the author is most known, whether that be his real name or not; hence entry under a steadily used pseudonym (as Mark Twain), under the maiden name of a woman till her married name is fully established, under a British nobleman's title by which he is always known (not under his family name, by which he is not known) is dictated by common sense. Of course this gives more trouble to the cataloger, and makes him sometimes inconsistent; but, as it certainly saves trouble to everybody else, and as the majority of the public do not care anything for consistency, it is the best practice. On the other hand, the cataloger, in his ambition to produce a fine piece of work, will spend hours in hunting up full names which when found are of very little use, if of any, the public. Understand me, in a large library, with numerous entries, author's names must be carefully distinguished or serious errors will result; but the town librarian, with ten or twenty thousand volumes, need not give every name in full, as if he were preparing a biographical dictionary. So in the early ages of cataloging in this country it was thought a fine thing to put in a brief sketch of each author - and, no doubt, it did good to the few who read the sketches; but to to It does not seem to me sensible to give up classification altogether, or even to give up minute classification because of its difficulties. They have been greatly exaggerated. The truth is that nine tenths of the books in a library treat of very limited, well-defined subjects, and so are easy to classify on any system - any reasonably good system; consequently they are easy to find on any system, and their classification is useful. The rest are troublesome; they have n't any very decided subject, or they extend over two or three subjects, or they contain several works bound together, or they treat the matter so that different classers would put them in different places, and if they are put in one place they disappoint somebody who would like to have them in another. I admit all this, but I say that this only applies to a small minority of books, and that the gain from thoroughly classing the large majority which lend themselves to classification far overbalances the harm that may come from the few refractory ones. In classification there has been a great controversy between the minute and systematic classifiers and the rough and ready classifiers. As I have been in the fight, it may not be becoming in me to assert on which side common sense stands. But at least I may claim that there is a certain plausibility in the following propositions: It takes time and costs brainwork to divide books minutely. It is not worth while to divide books more minutely than is needed for finding readily those which treat of each subject. When the subjects are distinct and generally recognized, setting each off by itself gives much more help than when they are vague and ill-defined. Therefore the sciences dealing with concrete matters lend themselves to minute subdivision much better than the abstract sciences like philosophy. Libraries used by the learned can profitably be much more subdivided than those used by the ignorant, who would not understand even the names of the divisions. When libraries are used by the ignorant and learned together, like college libraries, we must remember that the ignorant, especially if young, can be instructed. Libraries that have many books must be more divided than those that have few. Libraries that are going to have more books need to plan their classification with a view to enlargement as much as their building. These statements seem to me reasonably certain, and the practical result to which they lead is this: to secure the best results and greatest economy, a system of classification and notation should be contrived that in its simplicity and intelligibility will suit a very small library, the divisions being broad, wellknown subjects, and only a few subdivisions being made here and there of subjects on which the number of books is sure to be great, as fiction in literature; and yet these divisions and subdivisions should be parts of a great scheme that would suit a very large library, so that when the library grows and the different subjects increase -as they always will-irregularly, each can be subdivided when its time comes without disturbing any of the others and upon a plan prepared at the very beginning. This I believe to be more in accordance with common sense than to divide a little library with a minuteness that would suit the largest, or to consider a classification as a suit of clothes that must be entirely abandoned when the boy gets too large for them. The classification that I would have is the tree's bark that grows with it. I have praised common sense. But the librarian is no worse if he combines with it some grains of imagination and sentiment; if he is one whom the vast bookrooms of the great European libraries would strike dumb; if he feels an indescribable pleasure in hanging over an old manuscript or one of the works of the first printers, with its sturdy paper and honest ink, black as when first struck off, and its curiously irregular lines of type; if he prefers a quaint old binding to a necklace of jewels; if the odor of a case of books just come from abroad more delights him than a garden of flowers; if to him his library is the pleasantest place on earth, and his work there the most engrossing, the most satisfactory, that he can imagine; if every detail is to him of pressing importance, and he longs to perfect every part as the poet polishes his verses, and the painter retouches his canvas; if, as he answers the innumerable questions of the ignorant-and the learned — he fancies himself like the guide on the Alps, the pilot in the harbor, the equal of the teachers in the schools, the professors in the colleges, yes, of the pastors in their parishes. All of these delusions if delusions they are — will not harm him, for they are not inconsistent with common sense. For some introductory remarks, see PROCEEDINGS (First session). OUR SOME LIBRARY SUPERSTITIONS. BY W. I. FLETCHER, LIBRARIAN OF AMHERST COLLEGE. UR President has reminded us that any discussion of a subject should begin with definition. What, then, do I mean by "superstitions?" I find the dictionaries too strict in their definitions, the word being confined by them almost wholly to religious applications. I must, therefore, make a definition for myself; and I will ask you to let me call a superstition any idea or notion which is held as a matter of belief, and which is based on authority and accepted without reason, or the application to it of that ground principle in all good work - common sense. yaluable as the famed one of Alexandria which fell a prey to the opposite element of fire, there can be little doubt that in Ur of the Chaldees, while Abraham sojourned there, books were plenty and libraries flourished. The father of the faithful may well be supposed to have had a well-selected private library among the effects which he took with him when he went out to found a new nation. Since the researches of George Smith and others in the valleys of the Euphrates and the Tigris have unearthed the remains of those Accad libraries of baked clay tablets, the different books being stated by their makers to be copies of then ancient works handed down by former generations, the marvel of supposed accurate oral transmission through centuries of the most ancient records of our race has given way to the less incredible, but not less interesting, theory of a succession of documents, going back to the very fountains of the history of man. I have so often found myself in the position of a conservative as against many of the notions which have prevailed among librarians in recent years, that I take special pleasure at this time and in this live Western atmosphere, in attempting a little iconoclastic work. Because I desire beyond all else to be reasonable, I will go full length with any one against the twin irrationalities of an undue reverence for the ancient and an overweening ardor for novelties. I look back upon the progress that has been made in adapting libraries and library methods to modern demands and circumstances with a profound sense of satisfaction and a high appreciation of the labors of those who have been leaders and pioneers in this movement. But I am a béliever in evolution, rather than revolution, and when I compare the fruits of the two processes I find my faith justified. The library is an old institution. Our good brother Richardson, whom we all miss to-day, has a lecture on "Libraries before the Flood," in which, with the great erudition so characteristic of all he undertakes, he arrays the testimony of numerous writers of that early period, and makes a remarkable exhibit of the bibliothecal advantages possessed by the antediluvians. But whether or no it is to be believed that the waters which floated Noah's ark destroyed libraries as extensive and as If such be the age of the library as an institution, what wonder if, like other ancient establishments, it has become well encrusted with superstitions, or that some of these have become so firmly fixed in the very warp and woof of the fabric that they seem a part of it, and cling with the utmost tenacity even in the broad light of to-day? To begin with the enumeration of them, we will ask what notions with regard to library buildings may justly be considered as superstitions. In the first place, there's the sacred style of architecture, with its lofty and capacious interior, into which a chastened light feebly struggles from narrow windows piercing thick walls or from a few skylights in the roof. In the presence of American librarians of to-day, this superstition need not be dwelt upon. It is for us a thing of the past. But who can tell how many of us may yet be called upon (as was one of our number within the year) to try to administer a modern library in a magnificent new building erected on this old conventional plan? Just so often and so far as we can, it devolves upon us to denounce this superstition, and endeavor to create a sentiment with regard to it which shall reach and affect the building committees and architects who will yet be erecting libraries with one thought of the present and future and ten thoughts of the past. But while speaking of library buildings, I wish to indicate two other notions quite prevalent about them which, while not old, seem to me to be properly but superstitions. First, that of excessive regard for fire-proof ness. Books, pamphlets, and papers are inflammable to a high degree; and, while they are not rapidly consumed by fire, their backs, as exposed in a library, shelf above shelf, offer a ready food to the devouring element. This being so, there is but little security against a library's destruction by fire in metal shelves or uprights. Should fire once take in an iron stack of several stories in height, with perforated floors, I fear it would spread as quickly, and do as much damage, to say the least, as in a lower room with wooden shelving. It is conceded that the iron uprights are much more expensive than wooden ones of equal or at least sufficient strength. But the iron ones give an appearance of security against fire, and are often lauded on this account. This I call a modern superstition in library architecture. A library building ought to be fire-proof in so far as the structure of the building itself is concerned. But the book-shelving, being a mere shell filled with combustible material, can gain little, if anything, from being itself incombustible, especially, as in case of a hot fire, as much damage may result from its warping as from the burning of wooden shelving, or even more. The second modern superstition to which I wish to refer as connected with library architecture is the idea of making available for book-storage every perpendicular foot in the building. Certain librarians and architects have fallen a prey to this superstition, and seem to have become infatuated with it. It is the great central idea of the stack system. But it is not well grounded in reason. Why are not other kinds of buildings amenable to this principle? For instance, why does not a factory building fifty feet high to the eaves have seven floors? or a dwelling-house have only seven and a half feet between joists? Simply because the gain in such a method of construction would be more than offset by the loss. Higher rooms mean better light and air. Nor does the perforated floor make much difference. One defect about the stack system has lately forced itself upon my attention as it had not before. The roof light, when it has gone down two or three stories, disappears, and greatly to the relief of him who finds that the roof heat is also tempered as he descends. From this point down, lateral light must be depended on, and the attempt to force this side light into the interior of a library between floors seven and one-half feet apart means the employment of large windows and the admission of light and sunshine to an extent which is excessive near the walls. Books, especially the bindings, but also the paper, suffer from this excessive sunlight and sunshine to a very considerable extent. In short, the stack system is a strained effort (and an uncalled-for one, because where land is expensive elevators may be made to multiply the size of the lot to meet any reasonable requirement) to carry out against common sense and reason this mere notion (superstition, I call it) of "every perpendicular foot occupied." Not to dwell longer on superstitions connected with library buildings, I will mention some of a different class. First, there's the idea that a library must not part with anything which has once formed a part of its collections unless it be a duplicate. I dare say I shall here run counter to the feelings of many of my brethren, but I must maintain that there is an apparent lack of reasonableness about this notion of the sacredness of everything once in the library, so that it would be a sort of sacrilege to part with it. Looking into the near future, are we not led to the conclusion that our libraries must come to the point of a healthy sloughing off of the outgrown and obsolete accretions of the past, to make room for the constant addition of that which is vital for the present and the future? One other thing is pressing upon us in the East, where considerable libraries are growing up in almost every town, and that is the necessity and advantage of a differentiation of libraries, one following out one line of development and another a different one, and all helping one another instead of being engaged in a short-sighted rivalry. one fair way to set a price, and that is to With the acceptance of these two ideas,— that of keeping down the size of a library by getting rid of that which is useless and obsolete, and that of a harmonious and mutually helpful differentiation of libraries,comes in necessarily the abandonment of the old rigid rule of "Get all you can, and keep all you get," which seems to have prevailed hitherto, and which I do not hesitate to stigmatize as a superstition. I confess I have not yet seen indications of any decided escape from this superstition on the part of those having libraries in charge, but I expect to see them in the future. Reason will prevail here as elsewhere. Another superstition is the worship of decimals. I had the pleasure formerly of the acquaintance of an army engineer, General T. G. Ellis, who was a decided and earnest opponent of the metric system. I recollect a conversation in which he said that one of the great difficulties in the way of the progress of civilization was that the primitive man counted his thumbs. Had he only stuck to a truly digital system, we should have had a perfect method of reckoning. But as he was so unwise as to bring in the thumbs, we are saddled with a system of tens, in which a larger unit can be divided by two only once without a fraction. By the octal system three such divisions give us the lower unit and no fraction. The issue of the conflict between the artificial system of tenths and the natural one of halfs, quarters, and eighths, who can tell? At any rate, Mr. Dewey will tell you this is a very hackneyed and puerile argument against the metric system. It is not presented here with the view of running amuck against that system, but simply as illustrative. Perhaps we may be allowed to remark that no interest of modern civilization would have been more the gainer, had our early ancestors not counted their thumbs, than the one we represent. Right in this connection we come naturally to another superstition; namely, that of exchanging duplicates. We have had a great deal of talk about this matter of exchanging duplicates, and one proposition after another for a "clearing-house for duplicates." I have become satisfied that the best clearing-house for duplicates possible already exists in the form of the auction-room. And the only reason we have not all taken advantage of it is this mere superstition that a library ought not to sell for money what it may have to dispose of, but must exchange it for an equivalent. Now one of the greatest difficulties about an exchange is the fixing of prices. Whoever has tried it must have felt that he was put in a difficult and trying position. It is conducting a matter of trade outside of ordinary methods of trade. Prices on goods are supposed to have a definite relation to market value, as fixed by manufacturer or established by competition. But a great many library duplicates are not current in the market, so as to have any established price; and for all such there is but Since the days of Dr. N. B. Shurtleff, and his book on the decimal system as applied to libraries, we have been afflicted with a succession of efforts to run libraries on that system. Ten has been the sacred number with many librarians, as seven was with some ancient nations. There is something fascinating about the decimal system, it must be admitted; but it is when one is scheming on paper that this fascination is most strongly felt. In actual work a revolt against its artificiality almost inevitably sets in. By Dr. Shurtleff's method each range of book shelves was regarded as having ten shelves, each alcove ten ranges, and a room was naturally |