necessary qualifications for a children's librarian?" "Where can I learn landscape gardening?" "How can I prepare myself for settlement work?" are illustrations of requests received.

Another phase of the work is represented in the "Notes on women's vocations," which is one of the features of the Union house organ. These include references to new lines of work for women, openings in the Government service, civil service examinations, and notices of current books and periodical articles on vocational subjects.

In addition to the reference lists above mentioned, several special studies on vocational work with women have been prepared. A survey of the opportunities in eastern Massachusetts for professional training for women was made for the Association of Collegiate Alumnæ, and is embodied in the recently published directory of that organization on "Vocational training." A study of vocational guidance for college women has just been completed. This includes organizations engaged in the work, collegiate appointment bureaus, work of the Association of Collegiate Alumnæ and its branches, and college publications dealing with the subject.

ETHEL M. JOHNSON.

of paper and publisher's bindings, or even the unwarranted trimming and wholesale treatment that is characteristic of many socalled "library" binders, for the first we are unable to control, and the second may be regulated by the scheme outlined herewith to such a degree that the element of dissatisfaction may be greatly minimized, if not entirely obviated. Scientific direction makes it possible to use to advantage many of the methods that at present we are almost inclined to label as "butchery" and "shoemaking." Laced, or French joint cases, tight backs or loose backs, sewing on sunken cord, sewing on tapes, straight sewing, and over-sewing are all meritorious processes, but the exclusive adaptation of either alternative is a disastrous rut for

any library to follow. Yet that is pretty much the situation to-day as revealed by the bindings on the shelves of a dozen of our leading libraries.

STANDARDIZING MATERIALS AND METHODS

The number of colors for bindings should be few but strong in character, so that dark blue, for example, indicates definitely one exact shade, whether referring to a cloth or to a leather. Fancies or æsthetic delights deserve scant attention, if such selections suggest a multiplicity of rainbow effects, for thus the whole scheme of standardization is disrupted, since the term standardization includes the idea of perpetuity, and we cannot be sure that every shade can be produced or even matched indefinitely. The necessary idea of stability and the certainty that cloth has this lasting quality to a greater degree than any leather determines that the color scheme must have its basis in some one line of book cloths, such as the library buckram produced by the Holliston Mills.

For leathers a library should consider only two or three, namely: Turkey morocco, Niger morocco, and Scotch pigskin. Since the colors of pigskin are not dependable except in the naturals and the browns, this leather should be adopted only as supplementary. Contrary to what many leather dealers or bookbinders may say, perhaps because they do not handle the leather, it is

quite safe to trust to the qualities of Niger morocco. Turkey morocco has been the only leather to give uniform satisfaction for a long term of years, while other leathers have been only relatively satisfactory because of rapid disintegration after a short period of disuse of the books thus bound. The more recently introduced Niger morocco gives every evidence, however, of equalling the fine qualities of Turkey morocco, and costs several cents less per foot. Since the nature of this article will not permit any adequate discussion of the relative values of materials, for the purposes of this discussion, please accept the standardized selections herewith, made as a result of the suggested research and comparison:

With these colors of leather and cloth it is possible to match with reasonable satisfaction almost any number of old colors used, except for the old sheep binding. For this purpose natural pigskin serves best, and may be matched by Holliston Library buckram 396.

The style of type is another important feature to be considered. Too often have the binders been permitted to use their own taste, with the result that legibility is often interfered with. The old school binder practiced this little trick to insure the retention of the binding of future volumes of sets belonging to a library or to an individual patron; he would include in one title three or four styles as well as sizes of type. Frequently this result was a violation to artistic display as well as to the desirability of simplicity. Good type costs a lot of money (about eight cents per type letter), so that the judicious sprinkling of a dozen styles of letters over as many sets of volumes would be difficult for a better binder, perhaps, to match without a further equipment of type than would have been necessary otherwise.

The interior treatment and selection of

materials must be determined by each library and binder interested, although designed end-papers are recommended, since the figures hide the ugliness of the "turnin," the tapes, and the lining cloth. While not attempting to discuss the relative merits of case binding with the laced binding, it might be well to state that any volume bound in full cloth should never be of the "laced in" style.

CLASSIFICATION OF BINDING

An ordinary classification of binding is simply the division of new binding from rebinding, and while new binding, because it usually includes more difficult collation, averages a slightly higher rate of cost, this division is not important except as a matter of library record. A classification based upon the use, location, and value of individual titles, calling for four grades or qualities of work, is decidedly important. The principle of economy lies simply in the proper distribution of the total bulk of binding over those four qualities of work. Obviously, if the character of a large portion of the total binding of a library can be cared for satisfactorily at a cheaper rate than is absolutely demanded for a few volumes, it is a waste to select the superior grade simply for the sake. of having uniformly fine bindings. The four degrees of quality that are usually determined at a glance, together with the probable per cent. that each class forms of the library's total binding, may best be represented by the following outline:

Many public libraries might determine. that a considerable portion of class 2 could be dropped into class 3, because the general nature of the collection and the demands upon it would not require careful preservation of some bibliographic features that constitute items of distinct value in a university library. It is by no means uncommon, nor is it evidence of thoughtlessness or bad policy to find in certain library stacks valuable sets and works that have been bound very plainly, but in adopting this policy care must be taken that a cheaper binding for such material is not the sort that injures the books to the extent of robbing them of various bibliographic values. It has been determined by bibliophiles more or less definitely just how much worth may be attached to original bindings, uncut leaves, or any other items that may be characteristic of any particular edition. While this value is chiefly sentimental, the recognition of it has a slight economic bearing, in that any treatment which reduces that theoretical value also reduces the value of the collection. One authority states that trimming more than 1/64 of an inch from the top of a book takes off 50 per cent. of its edition value. If this assertion is true, there are many thousands of volumes that have been bound for various libraries within the last few years that are now about as valuable as a scuttled ship or a tumbled-down house. The universal practice of oversewing and its attendant processes by many binders is responsible for a great deal of this.

OVERSEWED VERSUS STRAIGHT SEWED BOOKS

Libraries owe a great deal to those binders who have developed methods of oversewing, because this process without doubt offers the most durable results. but this desired durability should be restricted to that class of books which demands it. In producing this binding the binders have made good, but the librarian has listened too long to their advice and has done too little investigation of the results if he permits the universal practice of this style of binding for his library. The process of oversewing usually requires the cutting off the back

of the books, taking away an eighth of an inch of the inner margin. The smooth back is then coated lightly with glue, and when dry, the book is separated into arbitrary signatures, which are perforated along a line an eighth of an inch further into this inner margin. After the volume has been sewed and has reached the forwarding process each of the other margins is trimmed smooth. The page has now been trimmed on all four edges, and because of a usual slight unevenness in the perforation and the nature of the sewing involved the edges of the sections are not aligned as evenly as is common with straight sewed books. For that reason the trimming of the outer edges is nearly double the amount that would be necessary to give the straight sewed book the proper finished appearance. The straight sewed books require the preservation of the original signatures, the number of pages of which are supposedly consistent with the weight and quality of the paper, while the oversewing process permits, at the point of the arbitrary division into sections, a regular variation from that number. A twelvepage signature book may thus be redivided into sixteen or eighteen page sections, although theoretically this never happens without intention. Whatever discussion may arise about this point, libraries may be sure that the usual result is a gain in the strength of binding, but that added strength is not always needed, nor is it desired, if the volume is a valuable one. The oversewing process is characteristic of cheaper bindings because it is labor saving; the actual time for sewing takes longer, as the binder will state, but the difference is more than made up by the time that is saved in the mending and reinforcements required on the back of the signatures, if treated in the orthodox way. That is why so many binders prefer to oversew every book they bind. The conclusion or recommendation set forth, however, should appease both this type of binder and the fastidious librarian. It is as follows: Give to those firms that have the best methods of oversewing all of the library's binding that would fall in the class numbered 3 in the

classification given herewith. The more valuable material is probably worthy of a more careful treatment, and seldom demands any elements of durability not consistent with the most orthodox binding.

THOMAS P. Ayer.

A. L. A. BINDING RECOMMENDA

TIONS

SINCE there has recently appeared a statement from a library binder that leather was the proper material with which to bind all books, no matter how they were to be used, it seems only fair that librarians should know exactly the recommendations of the binding committee on this important point. The recommendations advocated by the committee can be summed up in three brief rules:

1. Always use leather on books which are to receive hard usage.

2. Never use leather on books which will be seldom used.

3. In case of doubt give preference to cloth.

It follows from these rules that fiction and juvenile books should be bound in leather, except in localities where experience has demonstrated that cloth is better. In view of the experience of many libraries during the last ten years, there is no doubt in the minds of the committee but that leather is best for such books, and that a good grade of cowhide is good enough for this purpose.

Reference books, especially those which are heavy, such as dictionaries, encyclopedias, etc., should, of course, be bound in leather.

It follows, furthermore, from these rules that practically all periodicals should be bound in cloth. There are very few libraries in which the use of bound periodicals for reference purposes justifies hinding them in leather. There may be a large use of periodicals as a whole, but the use that any one volume will have year in and year out is very slight. Since the cloth which meets the specifications of the Bureau of Standards has been on the market librarians have had at their disposal a material

which, in view of the tests made before the specifications were drawn up, can almost be guaranteed to last as long as posterity will wish it to. We know that cloth which is very inferior in quality has been on the backs of books for over seventy years and is still in excellent condition. It is reasonable, therefore, to suppose that cloth made according to these specifications will last practically forever in the temperate zone.

While we know this about cloth we cannot be equally sure that leather will last nearly as long. We know positively that leather which is not free-from-acid is sure to deteriorate under conditions which will be found in all libraries. We know that leathers free-from-acid will last much longer, but how much longer is a matter of conjecture. Furthermore, it has been discovered that in many cases leathers which have been advertised to be free-from-acid have been found on analysis to contain as high as I per cent. of free sulphuric acid.

There is no question but that a leatherbound book has a much better appearance than one bound in cloth, but in view of the facts the Committee on Binding believes that the use of leather, except on books much used, is to be strongly condemned.

national library, with the duty to acquire and preserve not merely the most comprehensive exhibit of the American press practicable, but the miscellaneous material in every other field which the ordinary libraries can not undertake, and precisely because such other libraries can not undertake it-in the national library a limitation upon the acquisitions in concern merely for the difficulties of administration would be foolhardy. Better to receive it and treat it broadly than to reject it wholly." There is, however, the problem of actual shelving. Within six or seven years it will be necessary to arrange additional accommodation in the northeast court-yard as has already been done in the southeast court. A stack constructed there would provide space for 800,000 volumes at a cost of not over $325,000.

During the past year two of the veteran employes of the catalog division, Louis C. Solyom and Steingrimur Stefánsson, have passed away. Mr. Solyom, who was a Hungarian by birth, had been connected with the Library of Congress since 1867 and was a linguist of exceptional ability. Mr. Stefánsson, who had been with the library since 1899, was its highest expert in bibliography. A third death during the year was that of James Quay Howard, since 1897 in charge of the "Congressional Reference Library," with his headquarters in the Representatives' reading room. Several divisions, notably those of classification, catalog, and reading room, have suffered seriously through the frequent withdrawal of assistants to library positions elsewhere, where better salaries are offered.

The appropriations for the library proper and the copyright office, including those for the care of buildings and grounds, were $592,585.94. This does not include an allotment of $202,000 for printing and binding. Appropriations for salaries were $384,389.72, and for purchase of books $98,000, which was exclusive of $2000 to be expended by the marshal of the Supreme Court for new books for that body. The increase of salary from $6,500 to $7,500 for the librarian was refused, as were increases

for the chiefs of the periodical and prints divisions, and a special appropriation recommended for the purchase of additional books and other material for the division for the blind, was not granted.

The net accessions of printed books and pamphlets for the year were 115,862; maps and charts (pieces) 6100; music (volumes and pieces), 39,167; prints (pieces), 10,749. A numerical statement of the number of manuscripts is not feasible. The total number of books now in the library is 2,128,255; maps and charts, 135,223; music, 630,799; prints, 360,494. While no large groups of printed books were presented this year, the aggregate of 11,256 pieces received by private gift testifies to the continued interest and generosity of thousands of individuals and unofficial bodies. In a widely different category, but in its actual significance truly literary, since it embodies a memorial of distinguished literary service, was the gift from Dr. Lawrence Heyworth Mills, now a professor at Oxford University, of a beautifully illuminated address on vellum enclosed in a casket of silver, presented to him by Parsi friends and admirers in Great Britain as a mark of their appreciation of his services to Zend Avestic research. With the manuscript records of the American Colonization Society, came also the society's special collection of printed books, numbering 745 volumes, 730 pamphlets, and 486 periodical numbers. These included not only files of the publications of the society and of its state auxiliaries-reports, periodicals, and occasional issues-but also numerous miscellaneous works relating to slavery, to the progress of the negro race, and to the Liberian Republic. Some 500 photographs, chiefly of Liberian subjects, and 77 maps were also part of the collection. The classes of literature that have received the most concentrated attention and the most important, if not the most numerous, accessions during the year have been art and architecture. Source material relating to the early periods of discovery and exploration of the western hemisphere was largely augmented by the acquisition of the manuscripts of Dr. Rudolph R. Schuller, the well