College and Research Libraries Those difficulties will be with us even after the war is over. The Union List will serve as an invaluable aid in locating the issues of those put> lications that have been filmed by various libraries and agencies. It is highly desirable that all the cooperative project's of filming the elusive material abroad have their results recorded in the Union List. In the introduction we find the statement: "It is expected to continue the publication of this catalogue, with the prospect of a cumulative edition when personnel and re- sources permit a thorough bibliographical re- vision and a more complete indexing." Here is a hope that we all wish to see fulfilled.- G. F. Sheph e rd~ Jr.~ head~ circulation d epart- ment~ and in charge of microphotography~ University of North Carolina Library. Stepping--Stones to Cooperative Cataloging Coop erative Cataloging Manual for th e Use of Contributing .Libraries. Library of Con- gress. Descriptive Cataloging Division. Washington, U.S. Government Printing Office, 1944· 104p. The Library of Congress has issued a small Coop erative Cataloging Manual which should help to facilitate present-day cooperative cata- loging. At the same time it may serve as a point of departure for a discussion of co- operative cataloging of the future. The Manual~ prepared principally by Helen B. Stevens, describes the procedures of the cooperative cataloging centered in the Li- brary of Congress. It tells how libraries should proceed in supplying copy for printing but does not attempt to explain the intricate art of cataloging. In a brief introduction the h'istory of cooperative cataloging in the United States is sketched , from Charles Coffin Jewett's Plan for St er eotyping Catalogues by Separate Titles in 1851 to the activities of the A.L.A. Cooperative Cataloging Committee, begun in 1932 and merged, in January 1941, with the work of the Library of Congress. The cooperative cataloging associated with the Library of Congress has been concerned chiefly with the analyzing of serial publica- tions and. the cataloging of foreign books. Lately the work has been expanded to in- clude the cataloging of American doctoral dissertations, the output of a number of uni- versity presses, and the official publications of some of the states of the union, while the cooperative cataloging of the acquisitions by · several federal libraries in Washington, be- gun in 1902, has been .continued. Copy for printing is thus of various types and originates from many sources. The Manual gives useful information as to the manner in which copy should be prepared by the cooperating libraries. A special section devoted to the preparation of authority cards is in severfll respects more instructive and detailed than the corresponding section in the 1941 preliminary A .L.A. Catalog Rules. The work at the Library of Congress is likewise described, details being given as to · the re - ceipt and revision of copy, leading finally to the printing and distribution of catalog cards. An appendix contains a list of the 365 li- braries that have participated in cooperative cataloging, a list of 615 reference books use- ful in establishing and verifying author head- ings, and a list of easily understandable abbreviations that may be used aQ.vantageously by both the Library of Congress and the collaborating libraries. The latter feature con- stitutes in large measure the realization of a project that has for years been on the agenda of the A.L.A. Division of Cataloging and Classification. The Manual emphasizes that with respect to entries the A .L.A. Catalog Rules are gen- erally to be followed, while "beyond the head- ing, uniformity of practice is essential only so far as filing and intelligibility of the cards are concerned" (page I 6). Nevertheless, we learn on page 20 that the title of the book, including name of author, edition statement, and imprint, "should be accurately transcribed . . . according to L.C. cataloging rules." Collation, series note, and full name note are likewise to follow L.C. rules, while subject headings should be assigned according to the L.C. List of Subject Headings. Standardiz~­ tion, thus, seems to be more thorough than at first suggested. This contradiction touches upon a point of considerable importance for the future of American cataloging. When the revolt against overelaborate cataloging took place at the time of the publication of the new A .L.A. Catalog Code~ a distinct tendency 188 COLLEGE AND RESEARCH LIBRARIES to disregard standardization of descriptive cataloging was felt, many a library proceed- ing according to its own idea of rugged individualism. It is obvious, however, that, although the specific phrasing of a note may be immaterial, some uniformity of book de- scription is necessary if cards prepared by one library are to be used profitably by others. If, for example, some libraries use a full collation statement while others list only the last page of the main group of pages, how is one to interpret the collation symbols appear- ing on any given card? The Library of Con- gress realizes this; although it pays lip service to the new individualism, it is hard- boiled when it gets down to cases. For the cooperating libraries which have to follow the L.C. rules the situation is not so simple. The rules have been in a constant flux, and their latest version is not easily verifiable, if at all. Caught in the maelstrom of conflicting opinions, the A.L.A., having for three years delayed action, has not yet decided whether it wants to sponsor a de- scriptive catal0ging code, pending the next move of the Library of Congress. Plainly, what is needed is standardization of descrip- tive cataloging by as many libraries as pos- sible, as soon as possible and preferably in as simple a form as possible. This becomes more obvious the more we consider the results of coop~rative cataloging to date. Cooperative cataloging has been a noble experiment but, it must be admitted, not an entirely successful one. Two facts stand out clearly. The output has been far too small and the cost far too high. During the ten-year period from July 1933 through June i943 only about sixty thousand titles were cooperatively cataloged, or about six thousand titles a year, figures that in the words of the Librarian of Congress "are far from impressive." 1 Cost As to the cost, a study made by John R. Russell in •19372 reveals that the administra- tive and editorial expenditures involved in preparing for the press the 19,473 titles cata- 1 An.nu.al Report of the Librarian of Congress for the FtScal Year ended June 30, 1943. Washington U.S. Government Printing Office, 1944, p. 46-47. ' 2 Russell, John R. Report on the Work of the Co- operative Cataloging Committee." Catalogers' and Classifiers' Yearbook, No. 7 (r g38), p. 69-74. ll;f A RCHJ 1945 loged under the project of the A.L.A.. Co- operative Cataloging Committee from 1933 through 1936 amounted to $30,621, or $1.57 per title. In other words, it cost more ·to initiate and prepare for printing copy for a single title than it ordinarily does to catalog a title. In spite of all good efforts, the cooperative cataloging project failed to be- come self-supporting. Clearly, anything that · can be done to de- crease the cost of editing the copy should be done. The M apualJ we hope, will help somewhat in eliminating costly correspondence concerning points that should be clear to every cooperating library, but still more would be accomplished if we could remedy the weak- ness the Manual reveals: the absence of clear- cut cataloging rules. Although lack of uniform rules for de- scriptive cataloging militates against com- munal cataloging, it does not necessarily follow that, having agreed on certain funda- mental rules for descriptive cataloging, we could not allow a fair .degree of freedom in applying these rules. The resulting minor variations would probably not cause more trouble than the variations we have been ac- customed to accept on L.C. cards seen through the press by different revisers. If this is correct, we might in the future conceivably dispense with central revision of cooperative cataloging copy, except possibly for the head- ings. Since, according to Mr. Russell's study, the revision of the 19,473 titles produced during the period 1933-36 cost $r9,19I, or almost $1 a title, very considerable savings should be possible if revision could be largely eliminated. Going a step further, we might ask: If libraries can accept certain rules for copy contributed ·to the cooperative cataloging project, why can they not use th~se same rules for materials cataloged for their own libraries? If they would do that, it should be possible to have locally produced catalog cards universally used in other libraries pro- vided they were made available through an exchange pool. Weaknesses The Library of Congress has for a num- ber of years operated a service through which cards contained in the union catalog have . been made available to other libraries by 189 photostating. This service, temporarily sus- pended because of the war, · had, however, certain Qbvious weaknesses: I. A separate charge was made for all search- ing for copy in the union catalog irrespec- tive of whether o.r not the searching resulted in the location of a usable card. 2. Many of the entries in the union catalog were unsatisfactory since it was found that in I938 "not over 40 per cent of the entries • . • [were] of any appreciable use to catalogers."3 3· The cost of a complete set of cards . con- sisting of positive photostats was expensive (about thirty-five cents for five cards, in- cluding searching charge) . . It would seem that these weaknesses could in ,a large measure be overcome: I. If a second copy of each card submitted by a contributing library were interfiled in the Card Division's master file with the Li- brary of Congress's own card, a separate search--costly and frequently without re- sult-would be eliminated, since the card would be located in the process of the ordinary checking of L.C. card orders. 2. If adherence to ·uniform rules were made a prerequisite for including a local card in the master . file, the cards supplied would be certain to . be of a generally acceptable standard. 3· If the most economical process of multiple card reproduction were used rather than photostating, which is expensive except when a single negative is all that is needed, it should be possible t(') cut the cost con- siderably. Until inexpensive facsimile reproduction on satisfactory card stock is available, mimeographing may be the answer. The University of Texas has esti- mated that mimeographing costs amount · to about seven and one-half cents for five cards,' while the University of California has found the expense to be in the neighbor- hood of ten ceqts. Allowing for somewhat higher rates at the Library of Congress and a small f fee covering administrative ex- penses and filing, it would seem that the present cost of supplying a set of five cards could be cut in half. If a system of this nature, modified to meet 3 Library of Congress. Cooperative Work of Card Division, Union Catalog, Cooperative Cataloging and Classification Service and A.L.A. Cooperative Catalog- ing Committee. Thir d Circular. June I938. Washing- ton, I938, p. 25.-This publication, although a fore- runner of the Cooperative Cataloging Manual, is not mentioned in the Manual. 4 Coney, Donald [and others]. Report of a Survey af the Indiana University Library. Chicago, A.L.A., I940, p. 96. the specific requirements of the Library of Congress and the collaborating libraries, were adopted, the output of cards would increase and the cost decrease-the two ·objectives before us. , It would seem, then, that we would be well on the way to abolishing the present indefensible duplication of cataloging. Cards Supplied by L.C. Briefly, the Library of Congress would supply three types of cards: its own cards; cooperatively prepared printed cards of wide interest, for which stocks would be kept in the Library of Congress ; and mimeographed or otherwise duplicated cards, produced from unrevised exchange cards of more limited interest submitted to the Library of Congress by a group of libraries agreeing to follow certain specifications. In the case of exchange cards emanating from libraries willing to print and keep in stock a sizable supply of cards, orders for such cards might, after having been checke'd at the Library of Con- gress, be forwarded to the card-producing library to be filled. This system, obviously, wemld fit in well with the recording aspect of the Metcalf-Boyd-MacLeish· plan of di- vision of fields of acquisition but could be adopted should this plan fail to materialize . . It is quite likely, however, that eventually . we shall have to do more drastic things than to expand the present cooperative cataloging program and to arrange for the exchange of locally produced cards. There is no reason we should not, simultaneously with meeting the immediate demands, work seriously towards doing on a national basis what Mr. Rider has suggested we do on a regional one, · publish "a continuously cumulative book cata- log" that would serve at the same time as a national union catalog and, through location symbols, as the main catalog of individua,l libraries.6 The first step in this direction would be the publishing of the National Union Catalog in convenient book form, with a typography easier on the eye than that of the Catalog of Books Represented by Library of Congres·s Printed Cards. It would be a fitting memo- rial to Mr. Jewett if, in 1951, we could 1 Rider, Fremont. "Real Cooperative Cataloging- the Concrete ApP.roach." Library Quarterly I 3 :99-I I 2, ,April I943· We need not think that the potentialities of microprint, so stimulatingly discussed in Mr. Rider's latest book, should make us abolish the idea of the union book catalog. · 190 COLLEGE AND RESEARCH LIBRARIES Initiate a work that wpuld constitute the fulfilment of a vision doomed to failure a hundred years earlier but realizable now, thanks to a century's progress in printing processes.-] ens Nyholm, librarian, North- western University. Dissertations ·of 1943"44 Doctoral Dissertations Accepted by A m'erican Universities,1943-44· (Number II) Com- piled for the Association of Research Li- braries. Edited by Edward A. Henry. New York, H. W. Wilson Co., 1944. 88p. This new list, the eleventh in the series and the fifth under the present editorship, again shows thoughtful editing and increasing use- fulness. In general arrangement it is similar to previous lists. The seven main subject divisions have been retained but with ' litera- ture and art now more appropriately· headed Humanities. A few changes have been made in the subdivisions. Metallurgy _has been moved from Earth Sciences to Physical Sci- ences, and Geophysics has been added to Earth Sciences. There is the usual author index. The impact of the war upon graduate studies is reflected in the carefully prepared preliminary tables and introductory material. The number of dissertations presented has again declined. This edition lists 2117, the lowest number since 1930 and one almost 40 per cent lower than the high figure . of 1941. A brief table showing the distribution by large subject divisions indicates the increase in studies in the physical sciences. Sixty-five titles, largely in chemistry, are withheld be- cause they are "secret war research." The most useful of the introductory tables will doubtless be the one showing the practice of publication and loan of dissertations, and the list of periodic abstracting p~blications. Although the practices of publishing and lend- ing are too varied to be tabulated in exact detail, these two should prove especially ·valu- able to librarians on the borrowing end of interlibrary loan. Study of the table show- ing the distribution of doctorates for the years 1934-35 through 1943-44 by subject and years and of the one showing their dis- tribution for 1943-44 by university and by subject, will reward anyone interested in the general trends of graduate work on this level or in the relative strength of the v~rious graduate schools represented. The necessity for timeliness precludes the - possibility of indicating in the annual issues notes regarding the actual publication of in- dividual dissertations. It is to be hoped that at some not too distant date, however, it will be possible to have a cumulative index which will not only pick up the necessarily omitted titles b~t also show when dissertations have been published.-] ean M acalister, reference assistant, C9lumbia University Libraries, New York City. Study of the Army Medical Library The National Medical Library: Report of a Survey of the Army M edi~al Library Fi- nanced by the Rockefeller Foundation and Made under the Auspices of the American Library Association. Keyes D. Metcalf, ] anet Doe, Thomas .P. Fleming, Mary Louise Marshall, L. Quincy Mumford, a~d Andrew D. Osborn. xvi, 94P· Chicago, American· ·Librar.y A~sociation, 1944. Though we in this country have done some notable pioneering in the development of our municipal libraries, we have been slow in applying the same concepts of administration and service to our national libraries. Indeed, MARCH, 1945 until quite recent times ·we have scarcely thought of ourselves as having any national libraries. Outstanding as it has been for many years the Library of Congress, partly by virtue of its name, has taken a long while to establish itself in our consciousness as the national library of the United States. For a similar reason, the Army Medical Library (until about 1936 called by the still more restrictive name, Library of the Surgeon- General's Office), the largest medical library in the country, was the Army Medical Li- brary to us and not the national medical library. The unfortunate result of all this 191