r
R E Q U I E M
POR

C A R D

THE

C A T A L 0 G S

A History and commemoration of card cataloqs
in the Arizona State university Libraries

by

KENNETH B. KNEPP

L

1990

-

REQUIEM FOR THE CARD CATALOGS

A History and comm.ememoration of Card Cataloqs
in the Arizona state university Libraries

Arizona state University's library card catalogs are no
more.
The fingers walking through the catalog trays have
yielded to fingers on terminal keyboards. The familiar cards
have been replaced by displays on CRT's. The catalog records
tediously typed, or printed, or photocopied have given way to
the bits and bytes of the MARC records in the whirring disk
drives. After ninety years of faithful service, the cards and
the cabinets are gone. May they rest in peace.

Beqinninq9 of a Card Catalog
The first catalog card seems to have arrived in the Normal
School of Arizona Library at the very beginning of the century.
Gertrude Muir's history, The Arizona State University Library.
1886-1969, tells how, in March of 1900, Alma M. Davis, Student
Librarian, began classifying the books and preparing an author
catalog. 1 The inspiration and supervisor of this effort was
Professor F .M. "Cap 11 Irish, 11 f irst to attempt bringing order
out of chaos in the unorganized, uncataloged book collection
then housed in a large front room on the third floor of Old
11
Main."
Samples of supplies and of the proper catalog cards
[emphasis mine] were obtained [by Irish] and a copy of the
Dewey Decimal Classification purchased. 112
This catalog was begun for a collection reported in the
School's 1900 catalog to be more than 1200 volumes. 3 It seems
to have been a period of rapid growth as the catalog for the
previous year reports the collection at more than 700 volumes,
and the following year's figure is JOOO. In the 1904 catalog of
the Arizona Territorial Normal School, Librarian Peters reports
that a classed catalog of the book collection was available, in
addition to the dictionary catalog then being completed. When
she left two years later, the Library was said to consist of
5,ooo volumes.
The Normal School catalogs for several years
during this decade have the statement:
The library is classified according to the most
approved
methods
of
library
science.
A
dictionary
catalog
arranged
alphabetically
by
author, subject and title, greatly enhances the
usefulness of the library.

2

Information about the catalog over the next 25 years is
sketchy.
The first 1 ibr ary instruction class in 191O/11
included in its description 11 the use of the card catalog." A
report in the summer of 1910 counted 6000 volumes classed in
Dewey.
The library seems to have grown slowly, and it is
reasonable to suppose that the card catalog grew slowly along
with it.
The Bulletin of the Arizona State Teachers College at
Tempe for 1930/31, just before the Library moved into the new
Matthews Building, says that the accession record shows more
than 20, 000 volumes plus several hundred bound volumes of
magazines and several thousand unbound bulletins and reports.
There were also said to be 4 OOO children' s books in the
Training School collection.
Muir, however, has a figure of
14, 306
volumes
(excluding
documents)
and
2, 500
bound
periodicals moved into the new building.
The Librarian's
report to the North Central Association for that year states
the existence of a full dictionary catalog containing 46,550
cards (an increase of nearly 12,000 over the previous year's
report) figured by measurement at 100 cards to the inch. 4 It
may be supposed that this catalog moved with the collection.
The Bulletin for the following year reports the collection
at 18,000 volumes plus 2600 periodicals and several hundred
unbound items. The collection seems to have grown very little
during the depression years of the 1930's, being reported at
21,000 plus 4000 bound periodicals and 6000 public documents
for 1939/40.

Growth and Transformations of card catalogs
There is not a lot of information at hand about happenings
to the card catalog during the years in Matthews. It seems to
have moved about to different locations in the building during
two
major
additions
and
remodellings
and
various
rearrangements.
Apparently there were transformations of the catalog
during this period, but the clues discovered do little beyond
stimulating the imagination.
A Library Committee report,
undated but apparently from 1940 (last of T.J. Cookson's 20
years as Head Librarian), refers to
the confusion which results from having to
consult two card catalogs, an old one and a new one,
at least one of which is extremely unreliable.
The two gradually are being merged into one accurate
index, [but] the rate of progress is so slow that it
is difficult to predict when it will be a fait

3

accompli . 5
It then recommends either speeding the project with more
personnel or discontinuing it, merging the catalogs and weeding
cards for items not in the Library. No record of a decision on
this issue has been found, or of whether the intrusion of a
considerable war had any effect on these projects.
A following Librarian was in the middle of his 20 year
tenure when, in an annual report issued in January of 1954, he
mentions an accelerated recataloging project.
He further
reports the recataloging of 1, 519 volumes during the year,
along with cataloging 19,605 new volumes, and the entry of
68,658 new cards into the catalog. 6
The collections grew rapidly after the War with reported
figures passing 100,000 volumes in the early 1950's and
reaching half a million items by the mid-1960's.
In the year 1963 decisions were made which led to the
death of an old card catalog and its reincarnation in a new
one. The collection was to be reclassified from Dewey into the
Library of Congress Classification System.
The condition of
the old catalog determined that the project would be not only
a reclassification, but a complete recataloging. It would also
encompass a complete inventory and an evaluation and major
weeding of the collection The effort would take a dozen years
to complete.
There are several views of the state of the old Dewey
catalog before its transmutation. The report of a 1959 survey
of the Library by Richard Harwell and E.T. Moore for the
American Library Association states that "At ASU the purchasing
and cataloging abilities of the librarian himself have
apparently inhibited the development of strong acquisition and
cataloging departments." (This Librarian is reported to have
assigned Dewey numbers and done preliminary cataloging for
every item before passing it on to a cataloger to finish
processing.)
Harwell and Moore go on to recommend that
technical processing "should be in the hands of particularly
strong and capable department heads." Subsequently they state:
The card catalog of the Matthews Library is, despite
some minor flaws, in remarkably good shape.
As it
has been compiled wholly under the direction of one
chief
librarian
there
has
been
an
unusual
opportunity to maintain consistency in it . . . . The
very consistency of the catalog at ASU has created
some of its flaws.
Unnecessary detail (e.g. gift
notes, tracing information) is typed on the face of

4

the card.
An over-elaborate system of crossref erence cards has been begun.
Titles are
unnecessarily repeated in added entries for joint
authors and translators. Title cross references are
peculiarly used--in a way that too often makes it
necessary for a user to consult two cards before
finding the call number of the book he is seeking.
The catalog is cluttered by unnecessarily extensive
cataloging of government documents. 7
A 1962 report prepared for President Durham by P.M. Miles,
Assistant University Librarian, UCLA, also criticizes the
catalog for its excesses:
Too much attention is being given to cataloging as
separates
many monographic
series,
government
publications, and so on, which are actually more
conveniently
and
economically
handled
as
continuations.
With
the
indexes
and
bibliographies which are available today, no large
university library can afford to indulge in this
kind of detailed cataloging. 8
catalogers on the staff at the time, pointing out that
none of these outside critics were catalogers, were and are
inclined to be more blunt in their assessment of the old
catalog. The cataloging is described as old fashioned and of
very poor quality.
The cards are said to have been poorly
typed, half-spaced, crowded to near illegibility and to have
had corrections scribbled in pencil. The catalog contained a
lot of junk, and a great many cards for items that had been
gone for years and for ephemera that should never have been
cataloged. The handling of editions and series was erratic and
confusing.
The beginning of the reclassification project was followed
by a number of transformations in the catalog(s).
A new
catalog was started to hold all LC cataloging. In November of
1963 the Librarian wrote:
We will have two card catalogs for some years to
come.
We considered filing the two types of
cards into the same catalog but found it impractical
because the new, simpler f i 1 ing rules are quite
different from the system that has been used. 9
However, ten months later, he would write:
During the last three months of fiscal year 1963-64,
a crash refiling project was conducted to make the

5

Public catalog easier ... to use. The Dewey Catalog
was completely refiled, alphabetically word-by-word
and then the Library of Congress Catalog and the
Dewey Catalog were combined into one alphabetical
Catalog. 10
The Card Catalog in Full Bloom

June 15, 1964 was marked by an event with significance for
the future of the catalog.
On that date the first automated
system in the Library, an IBM circulation system, went up.
In 1964/65 an inventory was made, and in the following
year 70,000 cards for missing items were removed from the
catalog. Other major changes in the catalog also took place at
this time.
After 65 years as a dictionary catalog, it was
split into two sections, becoming an author/title catalog and
a subject catalog. In addition, "Clear labeling of the drawers
and of the cabinets was achieved [and] Many new guide cards
were added
and greatly facilitated the use of the
Catalog." 11
These latter accomplishments all coincided with another
momentous event, the move of the Library into the new Hayden
Building. The State Press for September 16, 1966 reports that
opening of the new building has been delayed for a week, and
that for a while students will have to use both buildings
because the books are in Hayden while the card catalogs are
still in Matthews.
Memories of people on the staff at that
time, however, are that the catalog had been moved during the
summer.
1967/68 brought another innovation designed to aid in the
use of this catalog. A special catalog Information Service and
Hotline were set up adjacent to the catalog.
This Service
featured direct telephone communication with the outside world
and with the floors of the Library and their subject reference
areas, eliminated trips to the catalog, and provided catalog
information to patrons and librarians. The State Press (Nov.
14, 1967) acclaimed it as a service not provided by any other
library in the U.S.
Meanwhile, reclassification continued apace. During its
first five years, through 1967/68, 72,833 titles were reported
to have been reclassed.
For the next few years, however,
little progress was made. Other projects and activities seem
to have taken priority, including addition of the millionth
volume in 1968/69. The reclass would finally be completed in
1975 with a final push to process some 25,000 titles.

6

William Axford, arriving as University Librarian in 1970,
brought several interesting innovations to the card catalogs.
The first and foremost was a further division of the
author/title catalog into an author catalog and a title
catalog.
This split was accomplished as a special project
during 1970/71. The author/title catalog before division was
estimated at nearly 1 million cards (10,000 inches) in 1,256
trays. An additional 680,725 cards (plus 48,575 order slips)
were reported filed in public catalogs during 1971/72.
With the author/title split came the decision to emphasize
the title catalog as the main catalog. Thousands of shelflist
cards were copied to make title cards for items which had not
previously been given title added entries.
These included
items for which title entries would have duplicated authors or
subjects in a dictionary catalog.
Also, holdings cards for
sets were to be attached to title cards, rather than to main
entries as they had been before. The quaint practice of tying
together entries with continuation or holdings cards with
string through the holes was also stopped at this time.
Not the least of the Axford changes was the introduction
of color into this three-way divided catalog.
Instead of
typing headings on added entry cards, the entry elements,
tracings titles, etc., were highlighted in color. The color
line was continued to the top left of the card to draw
attention to the filing element.
Green was used for titles,
orange for authors, pink for subjects and blue for series.
This procedure saved typing time, but added difficulties to
filing and possibly to the use of the catalogs. Another change
was that call numbers were no longer typed in the upper left
corner of cards, but were left where they already appeared on
printed cards, or typed at bottom left.
One more innovation in the same sequence was to declare
that the subject catalog would function as its own authority,
eliminating maintenance of a separate subject authority file.
Along with this, large numbers of guide cards (5 x 3 3/8 inch
cards with plastic protectors and with subject headings typed
at the top) were prepared and filed in the subject catalog.
In the spring of 1972 sample audits of the author and
title catalogs were conducted to test the accuracy of their
filing.
Error rates were found to be 3. 9% in the author
catalog and 5% in the title catalog. Nearly two-thirds of the
errors were minor errors unlikely to cause problems in locating
items. Most were the result of changes in filing rules rather
than original misfiling.
These results were substantially
better than reports from similar studies in other academic

7

libraries. A complete reading of the author catalog in 1972/73
found an even lower error rate of 2.6%. The author catalog at
this time was figured to contain about 920,000 cards in 1,052
trays.
The month of October 1975 brought two momentous events in
the history of card catalogs at ASU.
On October 31 the last
Dewey book in the Library was reclassed into LC.
A few days
earlier, OCLC had arrived and the first machine-readable
catalog record had been produced. OCLC brought the ability to
catalog books at a computer terminal connected by direct
telephone lines to computers and a huge,
fast-growing
bibliographic data base located across the country in Columbus,
Ohio.
It enabled ASU to share the cataloging of scores, and
eventually thousands, of other libraries. It made it possible
to push a button and order catalog cards which would arrive a
few days later, ready to file.
It also stored all of our
catalog records in machine-readable form creating great
potentials for future use . University Librarian Koepp' s report
for the year speculates:
It is even possible that OCLC may provide us with
the means by which it will be possible to give up
that marvelous 19th century invention the card
catalog, in favor of (a] more flexible and cheaper
microfiche catalog.
Still more basic, however
is the possibility of doing away entirely with
catalogs and depending upon terminals located in
public areas of the library to give location and
bibliographic information. 12
Aside from the implication that an online public access catalog
would be something other than a catalog, the statement may be
considered prophetic.
Catalog Cards

Having arrived at catalog cards produced by computer
controlled printers and shipped across the continent, this may
be an appropriate place for an interlude to think about how
cards have been produced.
This must depend a great deal on
inference and remembrance as documentary evidence is sparse.
The catalog cards which Alma Davis produced in 1900, and
probably those filed into the Normal School catalog for some
years to come, were undoubtedly hand written. The Library of
Congress did begin distributing printed cards in 1902, and this
eventually became the major source.
But, no record has been
found of when this library began acquiring cards from LC. The
Minutes of the School's Board record detailed expenditures for

8

the early years, 13 but some skipping through those for the
first quarter of the century found no payments to LC. They may
be there nonetheless.
There is no certainty, either, about when the first local
catalog records were produced on a typewriter. The Minutes of
the Board for October 23, 1924 show a payment of $70.00 to the
Remington Typewriter Company for a typewriter for the
library . 14
It cannot be said for sure that this was the
library's first machine, or that it was used to type catalog
cards.
It can be said with some confidence that, over the next
three decades or so, the catalog came to be composed of printed
cards with locally typed headings, and typed cards for local
cataloging.
Cards likely were acquired from H.W. Wilson and
possibly R.R. Bowker and other sources, as well as from the
Library of Congress.
The next major change in card production probably came
with the advent of the copy machine in the mid-1960's.
This
made it possible to type and copy unit cards locally, rather
than having to type complete card sets.
(Typed added entry
cards had been variously abbreviated.) Cards were copied onto
perforated sheets of card stock and then torn apart and
headings typed.
Memories recall the first Xerox machine's
habit of catching fire and filling Technical Services with foul
fumes.
The copy machines were followed shortly by the arrival of
LC proofslips, flimsy paper copies of all of LC's printed
catalog cards, supplied to the library on subscription. These
were filed by year and retrieved as needed to be used in
copying card sets. Assistant University Librarian Don Johnson
was responsible for another short-lived innovation in the late
1960's. This was a rig for using a 35 mm. camera to photograph
entries from the LC book catalogs and NUC's. The photos were
developed on campus, and copied to produce card sets.
The
results were very small print images on the cards, though a
Minolta copy machine with a specially developed enlarging lens
provided some magnification.
The general procedure of photocopying typed cards and
proofslips, varied by color-coding, persisted until the arrival
of OCLC cards in 1975. The Xerox 2400 machine which had been
used for card production was returned to the Xerox Corporation
on October 29, 1975. Color-coding of locally produced cards for
the East Asian catalog continued until mid-1988.

9

Decline and Fall of the card cataloqs 15

The reign of the card catalogs began to move toward its
end in 1979/80 with the decision to produce a COM catalog
(catalog on computer output microfiche). A strong motive for
this move was the plan to open a new Science and Engineering
Library in 1982 (subsequently delayed to 1983). It was clearly
not desirable or practicable to create card catalogs for the
new library and COM seemed to offer a good solution, at least
for an interim period. Other motives were supplied by a huge
filing backlog, major changes in cataloging practices with the
appearance of the second edition of the Anglo-American
Cataloging Rules and LC's decision to implement them and to
close its catalogs in the process.
The switch to COM for the new library required a major
retrospective conversion project so that all items to be
located in the new facility would have machine-readable records
and could appear in the COM catalog. About 75,000 titles were
converted between May 1980 and February 1981, mostly by AMIGOS
Bibliographic council, but with substantial cleanup by our
Catalog Service.
Special efforts by the Serials Unit were
required to finish converting science serials. The first COM
catalog was produced in the spring of 1981 by Brodart, Inc.
The filing of cards had ceased while the COM catalog still
in the planning stage. The official date for closing the main
public ca ta logs to new entries was December 1, 19 7 9 .
The
actual physical demise of the card catalogs began in the summer
of 1981 with a project to purge all cards duplicated by entries
in the COM.
All OCLC cards, all cards with science class
numbers (G-GC and Q-TX), and all "hard cards" (on-order slips
which for some years had been filed in the title catalog) were
pulled from the catalogs. Staff from all parts of the Library
participated in this effort.
It reduced the bulk of the
catalogs sufficiently for a consolidation in November of the
same year to empty several cabinets.
since it was not possible in 1982 to create machinereadable records with elements in nonRoman vernacular scripts,
access to materials requiring such records became a concern.
In that year cards were pulled from the main card catalogs to
start separate author, title and subject catalogs for ASU's
East Asian Collection.
Even while the first COM was in preparation, proposals
were being solicited for an integ~ated library system. Almost
as COM was being installed in the library,
DataPhase
Corporation was selected as the vendor to supply the integrated

10
system.
The online public access catalog component of
DataPhase's ALIS III system, however, was still several years
in the future.
The ALIS III OPAC actually became available to the
library's users in January 1986.
By that time, however,
DataPhase Corporation was in the process of selling the ALIS
III system to UTLAS Inc. which would rename it UTLAS T/Series
50. UTLAS' plans to support the system as it was being used at
ASU appeared problematical, so ASU spent fiscal 1986/87
shopping for and choosing a new integrated system. This time
the choice was the system offered by CARL (the Colorado
Alliance for Research Libraries) .
The Summer of 1987 was
devoted to converting and transferring ASU's bibliographic data
from one system to the other. ASU's CARL-based system became
operational in August of 1987.
The aspect of this activity most decisive in the demise of
the card catalogs was the conversion of manual catalog records
to machine-readable form so they could be added to the online
catalog. An effort to convert records for serials had begun in
1982 with the first of two LSCA grants for that purpose.
Serials retroconversion by ASU's Serials Cataloging Unit
continued over the next half-dozen years and was virtually
completed, except for a few "snags 11 , by mid-1988.
As the Libraries added their two millionth volume in
1984/85,
University Librarian Donald Riggs
received
a
commitment from the Office of the Vice President for Academic
Affairs for $1 million over five years for retrospective
conversion. A series of projects using the services of AMIGOS
began in the fall of 1984. Some three years and three projects
later, about 240,000 titles had been converted by AMIGOS and
ASU Catalog service staff. It would take many months more for
AMIGOS to produce the tapes of its output for input to ASU's
data base.
In 1988 ASU switched to OCLC for retroconversion
services.
Another 143,000 titles were converted during the
next year.
This project included the juvenile literature
collection, music scores, sound recordings, and twenty-seven
microf orm collections.
Almost 1 OO , OOO addition a 1 machinereadable records were acquired through purchase of MARC
cataloging for another fifteen microform collections.
As a result of these efforts, nearly all of ASU's catalog
records were in machine-readable form by summer of 1989. The
only significant exceptions were some records in non-Roman
alphabets and a number of microform collections for which
machine-readable cataloging was not available. The arrival in
1988 of the RLIN system for cataloging Chinese, Japanese and
Korean materials online made it possible to start doing new

11
cataloging for East Asian items in machine-readable form. As
this history is written, the retrospective shelflist for the
East Asian Collection has been shipped to a vendor for
conversion.
Unfortunately, the loading of some 130,000 retroconverted
records into the online catalog is being delayed by technical
problems and by computer programs which, in the fall of 1990,
yet remain to be completed.
However, by the time Reference Service moved to its new
location on the concourse level of Hayden Library in late July
1989, the card catalogs were largely redundant and were
abandoned.
This was in accord with library automation goals
and with plans for the underground addition and remodelling
which did not include space for the catalogs. ASU became one
of very few research libraries to have completely done away
with public card catalogs by 1990.
Within a few weeks the
author and subject catalogs had been dumped and the cards boxed
for use as scratch paper. The title catalog was subsequently
moved into storage adjacent to Technical Services for temporary
use in data base cleanup and maintenance projects.
Its
reprieve was a very short one, however, as by summer of 1990 it
was deemed to be unaesthetic and followed the others into
various forms of recycling.
The card catalogs have vanished from our midst.
"That
marvelous 19th century invention" served us long and well. Let
it be remembered with respect and admiration.
Kenneth B. Knepp, Catalog Librarian
October 1990
Thanks are due to Carl Cross, Betty Glover, Mary Kottke,
Clarice Mondo, Elliot Palais, Lois Schneberger, Mary swaty,
Jenny Williams, Ai-hwa Wu, Marilyn Wurzburger, and others for
their memories and records; and to the staff of the University
Archives and the Arizona Collection for help in locating
materials.

12

NOTES

1. Gertrude H. Muir, The Arizona State University Library, 18861969; a history {1971), pp. 3-4. Copies of this history are in
the University Archives, and in the Muir files in the Arizona
Collection of Hayden Library.
2. Unfortunately, neither of the two sources which Muir cites for
this information can be located in 1990. One was a letter written
by Davis to Professor John O. Grimes in 1950 and said to be in the
"Grimes file" in the Hayden Library Arizona Collection. The other
was information obtained by Ruth M. Wright, Librarian from 19111918, and apparently included in her scrapbook.
The scrapbook
according to Muir was in the Arizona Collection. It is now in the
university Archives [Ruth M. Wright. Tempe Normal School, Arizona,
Library (1918)) but does not contain the material or information
used by Muir. Board minutes do record authorization for Davis'
appointment and payment of his $10. per month salary through June
1901.
J. Much of the data on collection growth, etc. in this paper is
taken from issues of the Annual catalogue, Bulletin, General
catalog,
and
other
title
variations,
of
the
School/College/University under its various names. The University
Archives contain an extensive bound collection.

4. "Supplementary Report on the Library," (Nov. 1, 1932) [for the]
North Central Association of Colleges and Secondary Schools,
Commission on Institutions of Higher Education, [from] Arizona
State Teachers College, Tempe, Arizona, p. 2,
(Typewritten).
Arizona Collection, Hayden Library, ASU, bound in volume titled
Matthews Library. North Central Association Reports (Accession no.
90-337).
5. 11 Report of the Library Committee: to President Grady Gammage,"
p. [2], ASU University Archives, bound in volume titled Library
Report committee, 1937-1940, with the presidential papers of Grady
Gammage, Ninth President ... An inserted hand-written title page
carries the date: Oct. 4, 1940.
6. Harold w. Batchelor, "Arthur J. Matthews Library," Annual Report
of the President, Tempe, Ariz., Arizona State College (January
1954), p. 17 (Bulletin, new series, no. 7).

13
7. Richard Harwell and Everett T. Moore, The Arizona State
university Library, report of a survey of the library (Chicago:
American Library Association, 1959), pp. 12-14.
8. Paul M. Miles, "Notes on the Matthews Library, Arizona State
university, Tempe, Arizona: report to President G. Homer Durham 11
(1962), p. 12. Copy in University Archives bound with the Durham
presidential papers.
9. Alan D. Covey, Message from Matthews (Nov. 1963).
10. Ibid.

(Sept. 1964).

11. Covey, Alan D., Annual report, University Library,
State University (30 June 1967), p. 7.

Arizona

12. Koepp, Donald w., Annual report, University Library, Arizona
State University, 1975-1976, p. 9.
13. State Normal School at Tempe [etc.].
[Minutes], ASU, University Archives.
14. Ibid.,

Board

of

Education,

(Oct. 23, 1924).

15. Information about recent years is derived largely from the
memories and personal files of people still active in the Library.
Other data has been taken from Annual reports of the University
Librarians.