as delivered at the American Chemical Society Meeting, Chicago, August 26, 1993
There have been many cancellations of subscriptions to the
printed Beilstein Handbook of Organic Chemistry among U.S. aca-
demic libraries in the past two decades. As an academic chemis-
try librarian at Indiana University for nearly 18 years, the
author has participated in a number of surveys by the Special
Libraries Association and other professional groups designed to
determine which libraries continue to maintain subscriptions to
expensive reference sets. Each succeeding survey reveals a
dwindling number of subscribers to Beilstein. Cancellations of
Beilstein certainly took place in academic institutions before
the availability of the Beilstein Database. But, the fact that an
alternative "pay-as-you-go" source of information is now accessi-
ble online is seen in some quarters as a compelling argument to
cancel the printed subscription. Several librarians have pub-
lished in the library literature studies of the cost-benefit of
maintaining a print subscription to Beilstein. Invariably, they
have concluded that in their institutions, printed Beilstein
subscriptions should be stopped.(1)
The cost of a print subscription to Beilstein has risen
dramatically in recent years, over 50 percent since 1989/90.
1993/94 $32,041
1992/93 $31,743
1991/92 $31,351
1990/91 $23,945
1989/90 $21,185
1988/89 $21,764
Table 1. Print Subscription Payments for the Beilstein Handbook
at Indiana University, Fiscal Years 1989-1994.
By way of contrast, the full subsidy provided to the end-users
atall eight campuses of Indiana University for online searching
of STN International's CAS ONLINE Academic Program files has
risen 90-130 percent since 1989/90. In fact, CAS ONLINE costs
exceeded the printed Beilstein subscription cost in one of the
last three complete fiscal years, 1991/92.
1992/93 $27,203
1991/92 $32,953
1990/91 $22,177
1989/90 $14,295
Table 2. CAS ONLINE Academic Program Search Costs at Indiana
University, Fiscal Years 1990-1993.
It has been projected that the fifth supplement to the
printed Beilstein Handbook covering 1960-79 will be completed by
the year 2000. Furthermore, due to reduced production costs, a
prediction was recently made that the cost of the printed Beil-
stein will not be higher in future years than it was in 1993.(2)
The authors of one of the recent articles on the cost-
effectiveness of Beilstein conjecture that, "Neither library
staff nor the chemistry community may be familiar enough with
Beilstein to maximize its effectiveness."(3) If that possibility
exists for the printed Beilstein, the same could be true for the
Beilstein Database. The Beilstein Database currently has infor-
mation on over 5.3 million compounds. In late 1988, the database
was loaded on the STN International system, followed in October
1989 by DIALOG. In light of the availability of the online file
and the complementary Beilstein Current Facts CD-ROM product
which covers the literature from 1990 to the present, it was
decided to conduct a survey of academic libraries to determine
the level of usage of both the print and electronic products.
Since the author views the complete Beilstein product line as a
"database," the survey was not limited to electronic forms of
Beilstein.
Methodology, Population, and Sample
The lack of both time and resources to conduct a full-scale
scientific inquiry led to the decision to concentrate on the
academic subscribers to the Chemical Information Sources Discus-
sion List (CHMINF-L). CHMINF-L is a LISTSERV electronic mail
discussion group, which at the time of the survey had over 650
subscribers from a variety of organizations.
It is difficult to determine exactly how many of the sub-
scribers to CHMINF-L are at academic institutions due to the
idiosyncrasies of the Internet Domain Name System. However, at
least 125 different academic institutions were represented among
the subscribers on July 1, 1993. It was assumed that the chem-
ists and librarians using a source such as CHMINF-L are among the
more computer-literate potential users of the Beilstein Database.
No attempt was made to contact other categories of potential
users.
On June 1, 1993, a request for assistance in designing the
questionnaire was sent to 34 librarians who were CHMINF-L sub-
scribers on that date. These people were
selected because they were thought to be knowledgeable about
Beilstein, as judged from the quality of their contributions to
CHMINF-L or from past discussions of Beilstein with the author.
Three library school professors who teach science literature
courses were included among the 34 in that group. Seven re-
sponses resulted from the inquiry. In addition, a discussion of
the planned survey was held with six chemistry librarians on June
8, 1993 at the Special Libraries Association National Conference.
The feedback received from these efforts led to the development
of the preliminary questionnaire which was evaluated by a chemist
who has taught chemical literature courses. The final version of
the questionnaire was distributed via CHMINF-L
on July 7, 1993, with responses requested by July 26. After a
reminder was sent to CHMINF-L on July 19, a total of 35 usable
responses from 33 different academic institutions was received by
August 1. Five of the responses were from people outside the
United States.
Twenty five of the respondents were librarians, and ten were
chemists. Of those ten, eight were faculty members in chemistry
or a related science, one was a staff member in a chemistry
department, and one was a graduate student. Twenty one people
listed themselves as staff members in a library that serves
chemists.
Results
I. Use of the Beilstein Database and Current Facts.
Twenty six of the respondents had searched the Beilstein
Online Database in the past 12 months, with over half of those
being rather infrequent searchers. All but one of them performed
the searches themselves, and a number commented that the searches
were done as intermediaries for others.
6 times or less 9
1 time per month 7
2-5 times per month 7
6-10 times per month 2
> 10 times per month 1
----
26
Table 3. Frequency of Searching the Beilstein Database in the
Past 12 Months.
In about half of the institutions (N=16), costs of searching the
Beilstein Database are subsidized, but at eleven places there is
definitely no subsidy.
Since January 1992, academic subscribers to the printed
Beilstein Handbook of Organic Chemistry or the Gmelin Handbook of
Inorganic and Organometallic Chemistry or the Beilstein Current
Facts CD-ROM LAN version have been eligible for the "Springer
Chemistry Academic Program on STN."(4) The main features of the
program are:
- an 80 percent discount on the usage of the Beilstein,
Gmelin Online, and Gmelin Formula Index Databases
- a cap each year on the maximum payment of DM 10,000
per account number and per database
- database is accessible whenever STN can be searched.
STN is the overwhelming choice of vendor on which to search
the Beilstein Database at the responding institutions. Twenty
nine people search Beilstein on STN, whereas only six had
searched the database on DIALOG. Even those six indicated more
use is made of the STN system. However, only thirteen of the re-
sponding institutions receive the 80 percent discount, all of
them because of subscriptions to one or both of the print
products.
Significant numbers of cancellations of the printed Beil-
stein have occurred in the academic institutions in the past
decade.
Canceled in: N
1991-93 8
1988-90 4
1985-87 3
Prior to 1985 5
---
20
Table 4. Currency of Beilstein Print Cancellations.
The availability of the Beilstein Database was a factor in eight
of the cancellation decisions, but one person stated, "...online
availability was used as a 'ploy' to mollify faculty who might
have objected." Whatever the reason for the cancellations,
budgetary savings apparently resulted for all of the institutions
which canceled. None reported spending as much recently on
searching the database as was spent to maintain a full subscrip-
tion to the printed Beilstein, with one respondent indicating
"...less than $200 spent online last year, and probably much
less."
Interestingly, the availability of the 80 percent discount
for online searching of the Beilstein Database was not perceived
as an important factor for deciding whether to continue to sub-
scribe to one or both printed handbooks. Seventeen of twenty
four respondents said it was of little or no importance, six
considered it somewhat important, and only one described it as
very important. However, there is relatively little support for
continuation of the printed Beilstein after the completion of the
Fifth Supplement. Twenty one of thirty respondents indicated
they do not want the printed version to continue.
The Beilstein Database is not perceived as particularly
difficult to search. Sixteen of 29 respondents felt it was
either less difficult (N=2) or about the same level of difficulty
(N=16) as other databases. Three even stated that no databases
are difficult to search. However, ten people categorized Beil-
stein as more difficult than other databases with which they were
familiar. The great majority (21 of 27) think that the database
is less difficult to search than the printed Beilstein. If a
front-end software interface is used to search the database, it
is likely to be STN Express. Eleven of the twenty respondents
reported using STN Express.
The database is searched for a variety of purposes in aca-
demic institutions. These include reference, teaching, and even
as an index to the printed source. Respondents were asked to
rank their uses of the database using the following scale:
1 = rarely or never used for this purpose
2 = sometimes used for this purpose
3 = frequently used for this purpose.
Use Raw Score Total Score _N_ Average
3 _2 _1
ready reference 7 11 10 53 28 1.89
comprehensive data 6 12 10 52 28 1.86
teaching tool 2 5 19 36 26 1.35
rxn/preparation 10 11 6 58 27 2.15
other - - 9 9 9 1.00
Table 5. Uses of the Beilstein Database.
The data indicate that users of the Beilstein Database are fre-
quently seeking to find reaction or preparation information.
Somewhat less frequently, they are searching for quick access to
particular data or performing exhaustive searches for data. The
database is used relatively infrequently in teaching.
In comparison to other databases containing physical or
chemical property data, the Beilstein Database is clearly seen by
the respondents as superior. Eighteen of twenty six people
viewed it as more useful than other databases they had searched
for this purpose, three felt it was about the same, and only two
considered it less useful. Although the Beilstein Database was
not viewed by most respondents as overwhelmingly better than
others for reaction or preparation information, six of the re-
spondents believe it is better for this purpose, six feel it is
about the same, and six consider it less useful. The respondents
had searched a wide variety of databases in both categories, but
most frequently they were comparing Beilstein to the CA File: in
9 cases for physical/chemical properties and 17 cases for reac-
tion/preparation information. Next most frequently mentioned
were Gmelin (N=7) and CASREACT (N=6), respectively.
The capability to search the Beilstein Database by structure
is seen as somewhat important or very important by twenty two
people (11 each). However, seven consider structure searching of
the file to be of little or no importance.
For primary research published after 1979, the Beilstein
Database includes references to literature almost up to the
present. The entries for the post-EV data contain actual numeri-
cal values only for these five properties: boiling point, melting
point, density, optical rotatory power, and refractive index.
While all of these are considered by the respondents to be useful
properties to have in the database, they were less enthusiastic
about the last two. When asked to rank the properties on a scale
which had:
1 = of little or no use
2 = somewhat useful
3 = very useful,
the average scores showed boiling point and melting point tied at
2.54, with density a close third at 2.44. Optical rotatory power
scored 1.96, and refractive index, 1.69.
A minority of the institutions responding have made avail-
able the Beilstein Current Facts CD-ROM product; only ten of them
have it. Of those that do, eight of the ten keep Current Facts
in the library, one has it in an office, and another in a labora-
tory. Only three respondents felt that the CD-ROM has led to an
increase in searching the Beilstein Database. However, seven
believe it has increased the use of the printed Beilstein. There
is more use made of Current Facts than the database by six of
nine respondents, but also less use made of Current Facts than
the printed Beilstein by five of the ten respondents. Six of the
ten feel that the ability to search Current Facts by chemical
structure is very important, three view it as somewhat important,
and only one felt it was of little or no importance.
II. Use of the Printed Beilstein Handbook of Organic Chemistry.
A few additional general questions were asked about the
printed Beilstein. Earlier in this paper, it was noted that a
significant number (22 of 33) of the responding institutions had
canceled the printed Beilstein. It was of interest to determine
what other factors besides the 80 percent discount for searching
the Beilstein Database had contributed to that decision. Seven-
teen people pointed both to a need for savings to avoid canceling
primary journals and a perception of a very high cost per use as
factors leading to the cancellation decision. Other factors were
not considered important by many of the respondents. Only four
cited the availability of Beilstein at an academic institution
within a one-hour drive as a factor, and only one had gone so far
as to establish a borrowing arrangement with another library.
The sharing of such an expensive tool as Beilstein is typically
done by allowing people from outside the holding institution to
come for on-site usage, a practice described by seven respond-
ents. It was somewhat surprising to find that three institutions
were actually required by the administration of the library to
cut Beilstein. Among other factors cited for canceling Beilstein
was the long delay in publication of the data.
Eight of the respondents stated that their institution had
purchased the Beilstein Centennial Index covering the Hauptwerk
and supplements EI-EIV (literature through 1959) in a single set
of Chemical Name and Molecular Formula Indexes. Of those who did
not purchase the Centennial Index, cost was the over-riding
factor, although some cited the duplication of existing volume
indexes, and space considerations. There was also a feeling that
the set does not get enough use to justify the expense of the new
index. One confident librarian noted that "If you have a librar-
ian that knows how to locate compounds, the index is a luxury."
Several years ago the Beilstein Institute introduced an IBM
PC computer-based product to assist in finding the location of a
compound in the printed Beilstein. That is SANDRA, a tool which
allows the user to draw the structure with a mouse. In spite of
the existence of such a program, only eight respondents feel it
has increased the use of the printed Beilstein Handbook at their
institutions. One person noted, "...despite my efforts to pro-
mote it, few people ever use SANDRA." Perhaps some of the prob-
lem lies in the fact that SANDRA is only available for an IBM PC
or compatible computer. Fourteen people listed the Macintosh as
the computer they use most, with two of those also using IBM PCs
or compatibles. Twenty one respondents use IBMs exclusively.
Almost all (33 of 35) have their computers equipped with suitable
communications devices to search online.
The decline in the use of the Beilstein Handbook was noted
by twenty six of thirty two respondents, with only five believing
that its use has not decreased in the last 20 years. One person
stated, "I think the use of Beilstein has declined in the past 20
years as organic chemists have not brought along their students
as they were taught to use and rely on Beilstein. This product
has also suffered from the onslaught of online products/databases
that have spoiled chemists into an instant fix for information
which differs from the slow, methodical print-based research
performed by chemists only 10 years ago."
It is possible that the existence of central science li-
braries (or the inclusion of science materials in a general
library serving all disciplines) may have also contributed to the
decline in the use of Beilstein. Nine of thirty three respond-
ents indicated the library they normally use is in another build-
ing than the one in which they work. But it is likely that the
disuse of Beilstein could be linked to another factor not really
examined in this study--the relative decline in foreign-language
ability among scientists and librarians whose native language is
English.
There is also relatively less emphasis presently placed upon
formal instruction in the chemical literature by professional
organizations such as the American Chemical Society compared to
several decades ago. However, efforts are being made to teach
people how to use the printe Beilstein.
Although nine people responded that Beilstein Handbook users
at their institutions are self-taught, there is clearly an at-
tempt being made at most institutions to teach its use in a more
formal manner. People are usually taking advantage of the free
printed material available from the publisher (33 respondents),
but they also often produce their own materials to supplement the
publisher's materials (19 respondents). A few (N=5) even use
audiovisual materials.
There is considerably less training being offered to end-
users for the Beilstein Database. Thirteen librarians and five
chemists reported that they had the responsibility for offering
such training at their institutions, but an additional thirteen
stated that no one has this responsibility. In response to the
question whether usage of the Beilstein Database had increased
significantly after a training session or workshop was offered,
only three responded affirmatively. Six said "no," another six
said they did not know, and seven stated that there had never
been a training session.
III. User Comments.
A number of comments from Beilstein users were culled from
both CHMINF-L and this survey in order to lend a further dimen-
sion to this study. Ann Bolek of the University of Akron stated
on CHMINF-L,
I think that online searching of Beilstein on STN is very inex-
pensive, if you know what you are doing. I am usually looking
for one property for one compound. I crossover the Registry
Number from the Registry file to the Beilstein file, display the
fields available (D FA) and then display the information for the
property I want if it is in the fields available. My Beilstein
searches usually cost between $2 and $3, which is less than most
of my Chemical Abstracts searches, even with an Academic Dis-
count.(5)
An anonymous CHMINF-L subscriber stated:
I am a chemist, who does chemistry, and I often feel a little out
of place in this group which seems to be mostly chemists or non-
chemists who do library things instead of chemistry, but I'd like
to say something from a chemist's point of view. Beilstein is a
handbook, and it contains data. It also contains reactions, and
it contains references, but the thing Beilstein has that nothing
else has is data. If you need a melting point, or an nmr spec-
trum, or an ionization constant of an organic compound, there's
no quicker place to find it than Beilstein. If Beilstein is
infrequently used, it isn't because it's not useful, but because
chemists and the librarians who help them don't know how to use
it. In the last two months there have been a number of questions
posted here, for example, how to make chloroform from carbon tet,
where the Xray crystal structure of a phosphonic acid could be
found, what the ionization constant of an acid is, and several
others, and the answer was the same, it's in Beilstein. I am
fortunate enough to have instant access to part of Beilstein
(Hauptwerk through E-II), and not too inconvenient access to the
whole thing, and I use it at least two or three times a day,
often more. It's no more difficult to use Beilstein than thin
layer chromatography; all it takes is a little effort to learn
how, and like tlc, the more you use it the easier it gets and the
more you realize what it can do for you. The cost of printing
one copy of one subvolume of Beilstein must be about $50. It's
the cost of producing it (abstracters, computers,
subscriptions--the Beilstein Institute can't cancel Phosphorus,
Sulfur, and Silicon when it goes up to $7400) that makes it
expensive, and that's a constant. Each time a subscription [to
Beilstein] is canceled the constant has to be divided by a small-
er number. People who complain about paying $30,000 can thank
those who canceled at $20,000. Books are usually out of date by
the time they're cataloged, journal articles are available on
loan, but NOTHING can take the place of Beilstein.(6)
These sentiments were echoed by a respondent to the present
survey:
There is no substitute for Beilstein. The justification of
expense for dropping the subscription doesn't quite wash--Science
Citation Index is $15,000 for the CD-ROM, Beilstein is $30,000.
In 5 years SCI will be worth about zero; Beilstein will still be
as useful as ever. It is true that Beilstein is not used as much
as it should be, because the present generation of organic chem-
ists don't appreciate its utility. Part of the problem is that
the librarians don't know how to use it and don't appreciate its
utility, and don't suggest its use. STN's Numeriguide and CASRE-
ACT databases are beginning to do the same thing, but there's a
tremendous gap between E-V and 1992 when they start.(7)
Summary and Conclusions
Some academic librarians have chosen to sacrifice subscrip-
tions to the printed Beilstein Handbook of Organic Chemistry in
order to reallocate the money for other purposes. Since far less
money is typically spent on searching the Beilstein Database than
was paid for the printed Beilstein, money can be diverted to the
maintenance of primary journal subscriptions. There is not a lot
of searching of the Beilstein Database being done at academic
institutions, despite the existence of an 80 percent discount for
those institutions which do maintain subscriptions to Beilstein.
Structure searching capability is seen by most searchers as an
important feature of the database. Relatively few academic
institutions have subscribed to the Beilstein Current Facts CD-
ROM product. Those that have Current Facts report that it leads
to increased use of the printed Beilstein, something an earlier
computerized product, SANDRA, has not been very successful in
achieving. It was a bit of a surprise to find that Current Facts
does not seem to increase searching of the Beilstein Database.
A minority of academic librarians remain faithful subscrib-
ers to the printed Beilstein. However, there is a perception
among some chemists that neither librarians nor their colleagues
have either an appreciation for Beilstein or an understanding of
how to use it. It remains to be seen whether that same attitude
will be associated with the Beilstein Database.
REFERENCES AND NOTES
(1) See, for example:
Chrzastowski, T.E.; Blobaum, P.M.; Welshmer, M.A. A Cost/Use
Analysis of Beilstein's Handbuch der Organischen Chemie at Two
Academic Chemistry Libraries. Ser. Libr. 1991, 20(4), 73-84.
Knee, M. Beilstein and Gmelin: Keep or Cancel. Libr. Acq. Prac.
Theor. 1992, 16(4), 443-450.
(2) Luckenbach, R. Beilstein Centennial Index--180 Years of
Organic Chemistry (and 1.5 Million Compounds) at Your Fingertips.
Abstr. Pap. Amer. Chem. Soc. 1993, 205, Part 1, CINF 24. Com-
ments by Dr. Luckenbach to a question posed from the audience.
(3) Chrzastowski, Op. Cit., 82.
(4) Luckenbach, R. Beilstein at Academic Rates. CHMINF-L, Chem.
Inf. Sources Disc. List February 4, 1993.
(5) Bolek, A. "Re: Beilstein at Academic Rates." CHMINF-L (Febru-
ary 3, 1993)
(6) CHMINF-L: listserv@iubvm.ucs.indiana.edu
(7) Personal Communication