(1) For three subjects the oils of flaxseed, kaoline and peanuts,
there are titles listed under 665.3, but this class is not referred to on
the index cards for these oils; (2) for a number of special kinds of
vegetable oils there are index entries referring to 665.3, but no titles
for these oils in the classified catalogue; and (3) three index entries
are disclosed to be cross-references a negation of the advantage of-
fered by the classified catalogue of being able to use in the index any
number of synonyms for a subject as direct references to the appro-
priate section in the classified section of the catalogue.
The one thing right about the picture, and this a very important
Tightness, the general index entries for vegetable fats and oils re-
ferred to 665.3 where the reader finds, or should find, all monographs
in the collections which relate to this general class.
The ultimate solution to the primary wrongness of the picture can
only be, in terms of class notation, a specific class number for each
type of vegetable oil and fat. Another possibility is alphabetical ar-
rangement of the different oils under 665.3. This solution is effective
only in the absence of synonyms, except by the admission of cross ref-
erences in the index; but it does offer an immediate partial relief for
the difficulty. Still another possibility is to incorporate terms as mod-
ifiers of the class number referred to by index entries, in order to
eliminate the necessity of cross references in the index. If this were
done the index entry for Cocoanut oil could refer to "665.3: Cocoa oil
(Vegetable fats and oils):" and the specific titles on this subject
would be readily located by the subordinate guide card, cocoa oil, un-
der the general class guide card, 665.3 Vegetable fats and oils.
TEST CASE TWO
Here we examine a section of the classified catalogue in which
there has been completed an extensive revision. Prior to the revision,
plastics were classified under a general number for "gums, resins
and plastics," similar to the type of general class described under
Test Case One. In revising plastics were separated from gums and
resins and distributed under an expanded classification schedule.
In the alphabetical index to the classified catalogue, index entries
for subjects which have undergone such revision are on a different
colored card stock than older entries. The index now contains 125 en-
tries under Plastics or subdivisions of the general subject, referring
to 100 different class numbers in the classified catalogue. Four of
these sub- classes, in addition to the general entry Plastics, were
chosen for examination.
Plastics Accessory materials Solvents
678.042 (Plastics technology)
678.04 (Plastics technology)
Plastics Additives Dyes
678.047 (Macromolecular materials)
Plastics Additives Plasticizers
678.049 (Macromolecular materials)
1. 678 Plastics (Manufactures).
Under the general class number are filed eighty- nine titles.
Even here, there are some titles with special aspects presented,
such as chemistry of polymers or machinery for moulding plas-
tics. Cards for such titles normally appear, also, under other
classes. For example, some titles recorded here are also found
under 541.74 (Polymerism).
2. 678.042 Plastics Accessory materials solvents.
Two books (different editions of the same title).
3. 678.04 Plastics- Additives.
One book dealing principally with solvents. Probably, should
have been classed under 678.042. Some questions might be
raised on the appropriateness, either of the parenthetical ref-
erence in the index entry, or of the position of solvents in the
4. 678.047 Plastics-AdditivesDyes.
5. 678.049 Plastics- Additives Plasticizers.
In this test case we find that most of the deficiencies exposed in
Test Case One have been corrected. Particularly important, it is now
possible for the user to go directly from index entries to class num-
bers which cover only material related to the subject of the index en-
One of the aids for which we early felt a need was a manual of
practice for the construction and maintenance of the classified cata-
logue. We struggled with the problem for a time on our own, but
finally requested financial assistance for the project from the
Rockefeller Foundation. A grant was received, and editors were en-
gaged to write a guide. Dr. Jesse H. Shera and the late Margaret
Egan worked closely with the Crerar staff and the results of their
work appeared as a publication of the American Library Association.
It received one blistering review and a number of favorable notices.
It represents one stage in the work on improvement of the classified
catalogue which we were pleased to see realized. Its availability
eliminates the necessity of repeating here much additional informa-
tion needed fully to understand the characteristics and functioning of
the classified catalogue.
There are a number of additional comments which can be made to
throw light on the classified catalogue as a research tool, and on some
of the considerations which should be taken account of to make the cat-
alogue fully effective: (1) need for information on use of the classified
catalogue, (2) up-to-date classification schedules, (3) perspective of
the Library's holdings in any particular subject through the classified
catalogue, and (4) the high degree of expertness required of personnel
responsible for maintaining the subject catalogue.
I know of only one study which is concerned with determining how
the classified catalog is used by readers. This was conducted as a
master's study at the University of Chicago, by Emmett McGeever,
then a member of the Crerar Library staff. 5 He sought information
on the ratio of books found through the classified catalogue to total book
use, whether certain classes of readers used the classified catalogue
more than others, recency of items, extent of foreign language titles,
and extent of serials requested through use of the classified catalogue,
and what kind of difficulties were experienced by readers. The general
assumptions which he wished especially to test were:
. . . that the classified catalogue is used for subject access
to scientific and technical literature by the less experienced
searcher of the literature, who is not competent to take advan-
tage of the precision of the classified catalogue; and further,
that the use of the classified catalogue is a very low part of the
total catalogue use.
In the overall number of requests for books, 77.1% resulted from
use of the author-title catalogue, and 22.9% from the classified cata-
logue. This result appears to differ sharply from previous conclu-
sions that use of subject and author-title catalogues is about the same.
On the other hand, McGeever had anticipated an even lower percent-
age, and concluded that for even 22.9% of use of the collections to re-
sult from the classified catalogue was a very significant proportion of
the total use.
On the other hand, the first assumption which he wished to test was
strongly supported by the results, namely that the classified catalogue
is used by less experienced searchers of the literature. It was shown
that 14.1% of all books requested by subject resulted from use of the
classified catalogue by high school students. In reality, this percent-
age is much more significant than McGeever realized. During a gen-
eral study of reader use at Crerar late in 1958 6 it was learned that
only 3.8% of the total amount of reader use was by high school stu-
dents. This means that only 3.8% of total use accounted for 14.1%
of books requested through use of the classified catalogue. It is also
significant to note that McGeever' s study showed that 88.3% of the
books requested by high school students were located by the students
through use of the classified catalogue. In light of the further obser-
vation that all use of this catalogue was accomplished with very little
assistance by reference librarians, it seems justifiable to conclude
that the classified catalogue can be effectively used by readers rela-
tively unschooled in the technical aspects of catalogue construction.
We might glance briefly at some of the conclusions reached on
some of the other questions asked by McGeever as part of his study.
There was no significant difference in dates of publications of books
selected by use of the author-title catalogue and the classified cata-
logue. Of foreign language requests, 13.5% of all use was in this cat-
egory; but the comparable figures for the two catalogues was 16.1%
from the author-title catalogue, only 4.9% from the classified cata-
logue. An even more striking variance was shown for serials. For
requests originating from the author-title catalogue 86.9%; only 8.8%
from the classified catalogue.
Our interest here, however, is in use of the classified catalogue as
a research tool. For this reason, it may be of greater interest to look
at classified catalogue use by other reader groups. To refer again to
our 1958 survey of reader use, 7 we know that use of the collections by
the public is about equally divided between students, professional
groups, and technical employees of companies. Of non- student users
less than 3% are "general interest" readers. The major proportion
of student use, 82%, is by college and university students, with the
ratio between undergraduate and graduate students being approximate-
ly 3:4. Of non-student readers, the great majority are serious users
of the collections, again about equally divided, in this case between
employees of companies using the library in connection with company
business and professional workers in engineering, chemistry, medi-
cine and other areas pertinent to the scientific and technical fields
covered by the Library's collections. From these figures it is seen
that a very high percentage of book use by readers relates to serious
We might look, then, at one further result of the McGeever study.
While his analysis of use of the classified catalogue by groups other
than students was not extensive, he did find that the amount of such
use is very significant. For example, users of the catalogues engaged
full-time in use of the library located 29.4% of the books requested
through the classified catalogue. For those whose principal job is
library research, the percentage is 24.1%; and for those a minor part
of whose job is library research the percentage is 18.1%. And these
percentages represent use of the classified catalogue in its present
state of imperfection.
It is obvious from such evidence that we are justified in taking very
seriously the responsibility for making the classified catalogue the
most effective tool possible, but are faced by problems inherent in
accomplishing this objective.
Because of the requirement of specificity for greatest effectiveness,
it is necessary to have classification schedules which follow closely
the new developments in the subject matter of our collections. The
following information shows the universal character of this problem.
Problems of construction of the classified catalogue cut across all
disciplines within the scope of the Library. For example, the 1958
reader use survey disclosed a distribution of book use by broad classes
as follows: basic sciences 34%, technology 37%, medicine 26%, other
subjects 3%. Furthermore, active use by every reader group cut
across all three of the major sections of the collections. Only in anal-
ysis of subject use by particular professional groups do significant
differences become evident. A few examples from a separate analysis
of book use by professional groups are pertinent.
One might expect chemists to range widely among the subject clas-
ses, and this was shown to be the case. Of professional use accounted
for by chemists, the percentages for the three major subject groups
were basic sciences 26%, technology 17%, and medicine 6%. Engineers
and physicians would be expected to exhibit more specialized interests.
They do. Of professional use of medicine, 25% was by physicians, only
3% by engineers. Of professional use of technology, 36% was by engi-
neers, less than one percent by physicians. On the other hand, use of
the collections by lawyers, although relatively low, is significant, and
is about equally distributed among basic sciences, engineering, and
The pertinence of this analysis is that in the continuing review of
the classification and of the classified catalogue, no major discipline
can be slighted. The review must take place along the broad front of
all sciences and technologies.
Another problem derives from the need to have the classified cata-
logue serve well the value of presenting in perspective any given sub-
ject in relation to other sub-groups of any general class. One of the
conditions which makes this difficult is the large volume of material
which finds its way into the catalogue. It would be desirable, from the
point of view of this requirement to have numerous subjects repre-
sented on display by guide cards in any given catalogue tray. This is
often prevented by cards for one or two subjects so numerous that
they fill one tray and sometimes extend into another.
One possibility which occurs to us in this situation is to reduce the
number of cards. The feasibility of this is suggested by the natural
obsolescence of scientific publications in earlier years. The survey
of reader use, to which reference has been made, showed that cards
for earlier publications might be removed from the classified cata-
logue without materially reducing the value of the catalogue with re-
spect to coverage. For example, the statistics of use by date of pub-
lication showed that only about five per cent of total use of the collec-
tions was for titles published before 1900. And it is quite possible,
that further analysis of books requested through use of the classified
catalogue would show that imprints of much later dates for many sub-
jects could be removed. As a matter of fact, there is already some
evidence in the McGeever study to support this thesis.
For the purpose of presenting a perspective of any given subject
in relation to other sub-groups of any general class, another alterna-
tive is to print class lists for use of readers in different subject fields.
This has not been done at Crerar Library, but it is being given serious
Two related problems are of giving an overview of holdings relating
to a given industry, and an overview of the scientific and technical lit-
erature relating to a given region. Neither of these is adequately pro-
vided in the typical alphabetical subject catalogue. The first is only
inadequately provided in the classified catalogue. The major class for
technology of an industry has the primary material organized in its
various sub- classes, but related material in other fields can be traced
only through use of the subject index. The second problem of display,
however, is dealt with in the Crerar Catalog by use of appropriate
place numbers in the 900' s (not used at Crerar for general history),
followed by subject numbers, within parentheses, for scientific and
technical developments in the place or region.
Still another, and final, problem that has a high degree of relevance
to constructing effective classified catalogues is the heavy require-
ment placed on classifiers for subject specialization. It may be tnat
the solution to this problem is to draw into the classification activity
the expertese of more, if not all, members of the library staff. We
are exploring the possibility of this in our own organization. We have
about a dozen professional staff members outside the Catalog Depart-
ment who have sufficient knowledge of one of more areas in science
and technology to make a major contribution to such a program. And
we now have plans in the making to initiate a staff seminar on classi-
fication to explore the best procedures for utilizing this special know-
ledge to the benefit of the classified catalogue.
1. C.J. Frarey, "Studies of the Use of the Subject Catalog; Sum-
mary and Evaluation," Subject Analysis of Library Materials, ed.
M. F. Tauber (New York: Columbia University School of Library Ser-
vice, 1953), pp. 147-156.
2. Harriet Penfield, "Fifty Years of the JCL Classified Catalog"
(nine page manuscript in Crerar Library archives, 1952)
3. May I add a footnote here to the credit of Miss Penfield, whom
I greatly admired. Still alert for the late seventies, when she re-
tired, she was uneasy to see such a fundamental change in the charac-
ter of the catalogue take place, but readily agreed that it must be.
4. Jesse H. Shera and Margaret Egan, the Classified Catalog:
Basic Principles and Practices (Chicago: American Library Associ-
UNIVERSITY OF ILLlNUii
5. Emmett B. McGeever, "A Study in the Use of a Classified Cata-
log" (Unpublished master's thesis, Graduate Library School, Univer-
sity of Chicago, June 1928)
6. ''Survey of Reader Use of the John Crerar Library" (Chicago,
September 1958, Mimeographed report in process of revision) .
for the Reader
Director, Home Reading Services
The Detroit Public Library
Our library literature is replete with statements that indicate that
the goals and functions of the public library are vitally concerned
with the interests and needs of people in general. In fact our most re-
cent statement, as found in Public Library Service says in part: the
library's function "is to assemble, preserve, and make easily avail-
able to all people the printed and other materials that will assist them
Educate themselves continuously
Keep pace with progress in all fields of knowledge
Become better members of home and community
Discharge political and social obligations
Be more capable in their daily occupations
Develop their creative and spiritual capacities
Appreciate and enjoy works of art and literature." 1
Are classifiers and catalogers concerned with pronouncements
such as these? Or has it been assumed that a shelf arrangement
which stems from a classification which is a systematization of know-
ledge and originally was aimed at a service for scholars and special-
ists can logically be used by another service in libraries whose pur-
pose is primarily planned to provide the popular education services
for the general reader?
The well-established classification and catalogue departments in
large libraries make it seem efficient to class a book for a large
main library collection or for a series of special departments, and
then apply this same classification number for the book in branch li-
braries. This appears to be the quick and cheap way to do it. In the
smaller independent libraries the suggestions for class numbers made
by the H.W. Wilson Company, the A. L.A. Booklist, or maybe by the
Library of Congress, frequently aid the busy librarian to organize a
In 1937 in her book, The Classification of Books; an Inquiry into
its Usefulness for the Reader, Dr. Grace O. Kelley, originally a clas-
sifier and later a reader's consultant, highlighted the pitfalls of so-
called close or specific classification. 2 Those of you who remember
her book will recall her great concern with questions such as these:
Why do the methods of classification of books bring to light so small
a proportion of the library's total material on a definite subject? Or
why are the books on the subject in which one is interested scattered
in so many places on the shelves ? In speaking of her experience as
a reader's consultant in a large public library she states repeatedly
that to assemble books on subjects according to the way they had been
asked for by readers and according to the USE that was to be made of
them, it was almost always necessary to gather them from many places
in the classification system.
Visualize then if you can a non-fiction collection of 15,000 or fewer
books bearing class numbers which originally were assigned to books
in collections four or five times that large or for specialized depart-
ments. Imagine the kind of books that stand next to each other in a
bookmobile under even a fairly simple Dewey classification.
I think we will all agree that we have attempted to overcome the
hazards of shelf arrangements wnich stem from a Dewey-organized
collection by means of booklists, displays, more displays, exhibits,
etc. This can become what I call the bargain basement approach.
The English librarian Ernest A. Savage produced a whole book titled
Manual of Book Classification and Display 3 in which he has a chapter,
"The Home Reading Library as an Exhibition of Books", which pro-
vides detailed descriptions of equipment and methods for displays.
An interesting but not too helpful a book for us. We question seriously
whether these devices will ever be sufficient to overcome what may be
improper organization of material to begin with improper for accom-
plishing our frequently stated goals and objectives, these goals so suc-
cinctly stated in the Post- War Standards and again stressed in our new
guide, Public Library Service.
In the Detroit Public Library the responsibilities for these popular
education aspects of the service have been delegated to the Home
Reading Services as represented in the branch libraries and the
Home Reading Department at the Main Library. The special depart-
ments of the Main Library provide chiefly reference and research
services with a heavy emphasis on subject specialization. We in the
Home Reading Services also have a specialty it is that of being gen-
eralists. By that we mean that we have a working knowledge of good
books in all fields. We select our own materials and we believe we
need our own organization of those materials.
The Detroit Public Library's interest in a Reader Interest Classi-
fication goes back to 1936 when Mr. Ralph Ulveling, then Associate
Librarian, addressed a communication to branch librarians about an
experimental arrangement. In it he said:
For some time I have wondered whether our popular book lend-
ing service as organized on traditional lines is pointed directly
enough toward our service objectives; that is, whether the
organization of our circulating units is adapted to the function
we are trying to fulfill.
Interestingly enough it was in 1937 that People and Print by Douglas
Waples was published. In it he made the observation that a "psycho-
logical classification" of books was needed which would relate the pe-
culiar appeal that books make to readers of a given description to that
appeal. He also said that for a book "to pass from the stacks of a li-
brary .... to the reader's hands the publication must be advertised
.... It must escape the oblivion of the general catalogue or open
shelf and come to the prospective reader's attention."'
In his thesis on the purpose and administrative organization of
branch libraries, Lowell Martin repeatedly points to the necessity for
recognizing in the branch library different goals and functions than in
the main library. He says about the branch:
No longer can it look passively toward a continued role as a
poor copy of the main research and reference center. Rather
it faces a distinct task as an agency for distributing that por-
tion of popular educational, cultural, and recreational litera-
ture which the policies of the institution dictate .... The
branch is .... to be treated not as an imitation of the cen-
tral library differing only in scaling down of standards. 5
Later he says,
The number of potential general non-fiction readers may be
so great and service to them so important that a consider-
abele section of the book stock should be removed from the
conventional classification scheme and rearranged in terms
of the reading interests of this group. 6
I have used these fragmentary quotations to introduce the Reader
Interest Classification to show that our experiment is not merely a
scheme arbitrarily designed to do something different. It is a sin-
cere attempt to overcome hurdles in public service long recognized
WHAT IS THE READER INTEREST CLASSIFICATION?
It is a book arrangement that recognizes the variety of reasons
prompting people to come to the library. It is not a classification of