the scope and variety of the library's institutional public. The following types of institutions extend their local resources by borrowing from NLM:

Medical school libraries: Libraries of medical schools, colleges of medicine, schools of pharmacy, dentistry, nursing, osteopathy, veterinary medicine, public health, etc.

Medical society libraries: Medical and allied science association, society and academy libraries whether organized on a National, State, or municipal basis or by subject fields such as cancer, heart, etc.

Federal Government medical libraries: Medical libraries of any branch of the Federal Government.

Hospital libraries: Libraries for both professional staff use and patients' use in hospitals.

Industrial, research, and foundation libraries, medical and nonmedical: Libraries in both profit and nonprofit organizations in any field.

Other medical libraries: Libraries of State boards of health, city boards, etc.

International organization, medical or nonmedical.

For these libraries, the NLM constitutes a national resource, on which they can depend, and which they can use to augment their smaller collections, thereby guaranteeing to their users the availability of the scientific references desired. An anology to the Federal Reserve System is appropriate. Like the Federal Reserve Bank, the NLM is at the pinnacle of a national complex of local components. The Federal Reserve guarantees to local banks the security of their resources; the National Library of Medicine guarantees to local medical libraries ultimate access to the total resources of medical information.

These are the conventional channels through which the library conducts its national supportive mission. Each of these agencies, in turn, is concerned with one or more aspects of improving the Nation's health: the discovery of new medical knowledge through research, the improvement of health practices through the application of this knowledge, and education for the health professions. Only infrequently can any one of these objectives be isolated in terms of library functions. Usually, as in a medical school library, it is a question of degree of orientation of a library's loan, reference, and bibliographic activities to one of the several purposes. This is certainly true of the NLM, which, by virtue of the universality of its collections and services, provides national support for all three objectives.

The research process.-Scientific knowledge is accretional. Its records are built of reports of original investigations, the verifications, disproofs, and reverifications of these reports, their integration and consolidation into the preexisting body of knowledge. The published record of scientific research is voluminous, and of great complexity. In the case of biology and medicine, research is probabilistic in its methodology; this results in more attempts at verification employing differing methodologies, a process which swells the published record of the biomedical sciences to a greater volume than that existing in any other broad scientific field.

Three ingredients exist in the research process: the scientist himself, with his experience, his professional skills, and his curiosity; the laboratory, with its tools, facilities and resources; and the library, the custodian of the scientific record. The scientist uses the library's bibliographical tools for access to the scientific record, and its collections to consult the record itself. He needs to do this to see whether anything like his proposed study has ever been done, and if so, by what methods and with what results. He then conducts the investigation, using his personal skills and laboratory resources, and on completion of his work, publishes to disseminate his findings, to contribute to the record, and to deposit copies in the library. The library function is an essential part of the research cycle, without which the scientists cannot function, nor scientific knowledge advance.

NLM support activities.-The NLM functions to strengthen these processes through its various programs. For over 80 years the library has published the principal indexes for access to the research record of the medical sciences. In addition, specialized bibliographies, such as the definitive bibliography on Psychopharmaca, are published by the library to provide underpinning for emergent research fields. The Bibliography of Medical Reviews facilitates the access of researchers to that part of the published record which consolidates advances and developments on the frontiers of science.

Through its loan and photoduplication activities, the library acts to extend the resources of all other medical libraries. About 40 percent of the library's

ORGANIZATION OF PUBLIC HEALTH SERVICE

loans goes to Federal medical institutions and other Federal agencies, 20 percent goes to medical schools and societies, 20 percent goes to hospitals, and 20 percent goes to industrial, independent research, and other libraries. The significance of this activity is that it permits the alert scientist, practitioner, and educator, wherever he may be, to follow, study, and evaluate developments in his special field wherever they may occur throughout the world.

The library's translation activities are performed on behalf of large segments of the medical community. They aim, not at the researcher's ability to locate and acquire an obscure publication which may contain a highly significant finding, but rather at overcoming the serious inhibitory factor of his inability to read it, once found. After all, only 2 percent of all American scientists read Russian, and only one-half of 1 percent read Japanese. The distribution of selected translated papers helps to alleviate this problem.

III. TRENDS AND NEEDS

Science functions in an evolving society, and its institutional directions, practices, and forms are molded by the requirements of that society. Two major While observable in all developed countries, trends are dominant in our time.

they are particularly marked in the United States.

The publication explosion.-The first has to do with growth. Conscious of the success and benefits of research, society has increased its support a hundredfold. The great increases of dollars, of manpower, of facilities, and of research projects have been frequently reported to the Congress. Less well known is the effect of these increases on the scientific record, and on libraries. According to NIH data,1 each project grant generates one and one-tenth published papers; since 1945 NIH grants alone have swollen the scientific record by 50,000 articles. Comparable increases have occurred in all countries, resulting in what has been called the publication explosion, a phenomenon which has overburdened scientific publication media and libraries equally.

The interdisciplinary revolution.-The second has to do with a change in the organization of scientific research. Its 19th century form followed academic lines; institutes, laboratories, and careers alike were cast in the mold of disWorld War II demonciplines: e.g., microbiology, physiology, biochemistry. strated dramatically to the supporters of research that society could achieve scientific breakthroughs through interdisciplinary combinations, through the assemblage of research teams representative of multiple related disciplines. Common now are areas of science comprising 20 or more disciplines, associated for work on a problem complex, or a specific disease (e.g., space biology, diabetes, molecular biology).

This interdisciplinary revolution in the sciences has been accelerated by largescale funding; it has changed the organization of research drastically within a generation. In terms of the scientific record, its effects are profound. Each new interdisciplinary field, for example, has it own communication requirements: journals, abstracting services, bibliographies, critical reviews. Since the organization of the traditional scientific record followed discipline lines, the need to repackage the record has resulted in great dislocation: the fields staked out by journals overlap, new abstract services are generated to supplant timehonored ones, and there is a general insecurity about the adequacy and the permanence of the scientific record.

It is these two forces, taken together, with underlie the "crisis in scientific The national research effort is generating vast amounts of communication." new published information, and the traditional forms of organizing it for furthering the effort appear to be inadequate.

Impact on the National Library of Medicine.-These two circumstances have profound and far-reaching implications for the NLM, and for the medical library network with which it is affiliated. The world in which they had their historical development has been disoriented, and its axis shifted. The fundamental problem facing the NLM, as well as other institutions involved in the communications process, is to accommodate to this new orientation and new magnitude of research. The NLM's major response is the medlars system of computer-based indexing, previously described. The fundamental significance of the computer in a largescale information storage and retrieval system lies in its ability to manipulate With medlars, the library will have masses of material along multiple axes.

1 Lindsay, D. R., and Allen, E. M., "Medical Research: Past Support, Future Directions," Science, 134: 2017-24, 1961.

32-692-644

the ability to unitize new medical information, and to put these units together again in any pattern or combination which may conform either to disciplinary or interdisciplinary requirements. Medlars, in short, is the bridge between the two worlds, between the past and the future.

National significance.-Computer-based information systems developed or contemplated to date have been restricted to the use of a particular group or company, and are, in a real sense, private systems. The significance of medlars is that it has a potential for controlling the largest of the scientific literatures for one of the largest professional groups as a public service function. On completion, this computer-based indexing system will be a unique asset for American science. The library expects and intends to share this Government-owned asset widely with other institutions, thereby to realize as completely as possible this national potential.

"Current awareness” services. To keep scientists up to date on newly published research results published throughout the world, specialized rapidly produced listing services are an essential. With increasing frequency the library has been approached by interdisciplinary research groups, representing both public agencies and private societies, with request for monitoring devices which will cover the world's literature of special interest to them. In a first phase, the library plans to provide up to 50 current listing services, each tailored to the specifications of a research group, as one of the principal medlars products. Through the duplication of medlars tapes, and the use of computers at university centers, this facility can be progressively expanded in a second phase on behalf of still more groups. In a third phase, such listing services can be provided for research teams and individual laboratories in many centers.

Toward a national retrieval network.-The unprecedented search and retrieval capability of the medlars system can and should be widely duplicated if the full benefits to American medical science are to be realized. Following the principal that bibliographic access and physical access should go hand in hand, the NLM has developed preliminary plans for the establishment of a national retrieval capability in which medical libraries at university centers throughout the country will have a central role. Prelminary plans have been developed, which call for the duplication of tapes from masters prepared in the NLM, and their distribution to other libraries having access to computers. Much further developmental work must be done before such a system can be an operating reality, but the library looks forward to the day when 50 libraries in as many States will utilize the Index Medicus published in a magnetic tape edition for daily machine literature searches in support of local research needs.

Depressed state of medical libraries.—This unprecedented retrieval capability, however, has little meaning to the libraries or to the national research effort unless it is matched with adequate availability of the literature itself. For years it has been increasingly evident that medical libraries have been undersupported by the schools and institutions which they serve. Ten years ago, when the Federal funds for medical research were half what they are now, the professionally conducted survey, "Medical Schools in the United States at MidCentury," " found "little evidence of attempts to meet the libraries' increased needs during a period of tremendous expansion in the research and other activities of the schools." The report concluded that faculty workers, research workers, and students would be "severely handicapped unless the new demands placed on libraries are met by careful administrative and budgetary planning. In a Nation dependent on medical research to a greater degree than ever before, surprisingly little is being expended on the housing of the reports of that research and on making those reports available."

The Deitrick survey found the medical libraries had accumulated deficiencies through years of undersupport, and judged them to be dangerously inadequate to meet the needs of medical research and eduaction 10 years ago. In the 10 years since the report, wtih greatly increased burdens placed on them, the libraries have not held their own; research expenditures have increased eightfold; the amount spent on libraries but doubled. Of all the resources supporting national programs of research and training, it is obvious that the stored intellectual resources, represented by the medical libraries, is the weakest. Their weakness is evident in four aspects: books and journals, skilled manpower, outdated technology, and housing.

2 Deitrick, J. E., and Berson, R. C., "Medical Schools in the United States at MidCentury," New York, McGraw-Hill, 1953.

Library resources.-The demands of interdisciplinary research for library materials in basic scientific fields not previously considered to be within the scope of a medical library have caught many libraries unaware, and unprepared to acquire, process, and store the large amounts of new literature required. Library budgets, already inadequate for the purchase of books and journals in the conventional medical sciences, have been overpowered. The median medical school library of 53,000 volumes is about half of what it should be to support broad biomedical research programs.

Regional medical libraries.-Faced with these deficiencies, libraries in metropolitan centers have entered into cooperative agreements for the pooling of Several areas (New York, Boston, Pittsburgh, and Cleveland) their resources. have preliminary plans for regional medical library centers, which would store and service on behalf of the entire metropolitan library complex, the lesser used but scientifically important literatures of medicine and its related scientific fields. Dollar resources have been lacking to support this promising trend, which is entirely compatible with the NLM plans for decentralizing the medlars system.

Manpower deficiencies.-Manpower deficiencies are both quantitative and qualitative. A vicious circle of low performance requirements, low salaries, low training resources, low recruitment successes, and again low performance requirements has led to a depressed situation where able, qualified medical librarians are in distressingly short supply. The median number of librarians with professional training working in schools of medicine is 3; the total professional manpower pool for 80 schools of medicine is 326.

The revolutionary changes in the scientific bases of medicine, the new fields of interdisciplinary research, the rapid advances being made in the organization, storage, and retrieval of medical information have caught the medical librarians grossly underprepared. The talents, for example, to master the manmachine interrelationships of the proposed decentralized medlars system are almost nonexistent.

The strengthening of educational resources, the vigorous recruitment of new manpower, the retraining of librarians already on the job, are essential if the human resources necessary to the servicing of the information needs of medical research and education are to be assembled.

Outdated technology.-The technologies related to the provision of better bibliographic access (e.g., machine storage and retrieval of information) and to physical access (e.g., photocopy devices, facsimile transmission) have made giant strides over the last decade. Medical libraries have been slow to profit by these advances primarily because of the developmental costs involved, and the capital outlays, which, even though modest, have been beyond their ability to fund. If the libraries are to become effective instruments for supporting research and education, attention must be given to research and development to bring the advantages of modernized technology to the library process.

Construction needs.-Most urgent is the need for more and improved housing for medical libraries. The majority of libraries which were planned a generation ago were underplanned at the time, were hopelessly overcrowded 10 years ago, and are desperately in need of more space today. The "publication explosion" cannot be accommodated within the four walls of underplanned libraries. As a consequence, in many institutions, publications essential to research programs are warehoused miles away from the school campus, making the ready access of scientists to the literature a fiction. The operating inefficiencies of such medical libraries handicap severely the educational and research programs conducted within the schools.

3

Seventy percent of American medical school library buildings are more than A 1957 survey showed that over 10 years old; 50 percent more than 30 years. half of the medical school libraries reporting had exceeded their shelving capacity, and but a handful of new buildings have been achieved since.

Recommendations of National Advisory Health Council.-The National Advisory Health Council has passed repeated resolutions calling attention to the serious state caused by library inadequacies. Its resolution of November 1962 reads in part: "The medical library network which has been designed to make the published record of medicine available is in dire trouble. During a period of intensive development of research institutions, their essential library support has been seriously neglected. In recent years, the needs for adequate library

3 Fry, A., and Adams, S.. "Medical Library Architecture in the Last 50 Years." Bulletin. Medical Library Association, 45: 47-79 (1957).

working and storage space, for more trained library personnel, and for new methods of handling and disseminating the growing scientific medical literature have become acute."

Bloomquist survey.-A 1962 study of the accumulated needs of medical school libraries which is equally applicable to other types of medical libraries— iterates these deficiencies, points out that little improvement has occurred since the Deitrick survey of 10 years ago, and makes the following recommendations for an action program, involving Federal assistance:

Recommendations

1. Federal funds should be made available to both existing and new medical schools for the purpose of building up their collections to a minimum adequate standard. It is suggested that this standard in 1962 be 100,000 volumes and the current receipt of 1,500 journals.

2. Federal granting agencies should take realistic responsibility for the support of their grant programs in terms of the literature, library staff, and library services required, just as it does in terms of equivalent technicians and laboratory services required.

3. Federal funds should be made available for the support of the essential bibliographic apparatus needed in medical research. This apparatus today is breaking down. The National Library of Medicine should take national leadership in medical bibliography in its multidisciplinary aspects.

4. A system of regional reservoir libraries should be established in the United States, encouraged and supported by the National Library of Medicine through the granting of funds, the development of an improved bibliographic apparatus, and the development of an improved communications network.

5. If science information centers develop with the aid of Federal funds, maximum advantage should be taken of existing library resources in order to minimize useless and costly duplication.

6. Federal funds should be appropriated to initiate a vigorous program of recruitment, education, and training of medical librarians aimed at raising the standards of medical librarianship and creating a manpower pool. These funds should support and expand the Medical Library Association's present programs of scholarships, fellowships, internships, and continuing education. In addition funds should be made available to qualified academic institutions for the initiation or improvement of their training facilities in this area.

7. Federal matching funds should be made available for the constructing, equipping, and renovating of medical school library facilities under the research facilities construction program. Federal grants to new medical schools for library construction and equipment should be provided under the proposed educational facilities construction program.

8. A program should be established whereby nonrecurring Federal grants for special projects would be made available directly to medical school libraries. These would tend to be small in amount and for the purpose of catching up with existing backlogs.

9. Experimentation in new electromechanical and reproductive devices relevant to library problems should be encouraged and funded.

NLM interests.-As a national library, with a statutory responsibility for the dissemination of health and medical information, the NLM is deeply concerned with the health and service capabilities of other medical libraries. It is well aware of the national need for the strengthening of medical library resources and services. Even though the library had no national system for mechanized storage and retrieval in contemplation, a large-scale program to raise these libraries out of their present depressed state to a level adequate for the support of present-day research and education is still urgently required.

As the world moves into the computer age, and the medlars system becomes operative, the underdevelopment of medical libraries poses a critical problem. How will the power of electronic retrieval benefit local research needs if the local medical libraries cannot supply the literature retrieved? Of the two basic research library functions, the provision of bibliographic access and the provision of access to the materials, the former can be performed with a tenfold increas eof efficiency; the latter remains grossly depressed The mismatch of capability is great, and comparable to the installation of a turbofan motor in a model T chassis. The full benefits of the medlars to the American medical

Bloomquist, H., "The Status and Needs of School Medical Libraries in the United States," Boston, 1962.