The Creation of Canadian-Based Bibliographies of Early Books

itudes.' They are willing to entertain a wide variety of views about the nature of those two solitudes and to provide solutions to the problems posed, from a spirited bilingualism and biculturalism to a belligerent assertion of the value of unilingualism or an alleged superiority of one culture over another. But Canadians are locked into a complex nationalism and interdependence by their common concern for a large part of the North American continent and all that its common history stands for. It is not too far-fetched to assert that in the world of books and what librarians call 'non-book materials,' i.e., other records of intellectual and cultural activity of the human race, there are also two solitudes locked into a common interdependence. Natural languages do not divide the bibliographer and the librarian, but their common language (or perhaps 'jargon' might be a better word for it) is so heavily strained by their differing understanding of common terminology (even 'bibliography' itself) that one cannot but conclude that each belongs in a different solitude. It might be pushing similitude too far and too dangerously to assert that librarianship is the majority culture and, in Canada at least, that the majority culture has attempted to assimilate the minority one. The Bibliographical Society of Canada might seem to have adopted this solution to the problem, for one cannot but be struck by the fact that it alone among 'national' bibliographical societies is very much dominated by the majority culture (witness the Bibliographical Society of America, which keeps librarianship at arm's length). Librarians have been among the most active members of the Canadian society, and it traditionally meets annually at the same time as the Canadian Library Association. But recent attention to the concerns of 'analytical' bibliography rather than 'enumnerative' bibliography suggests that the minority culture is gaining a voice and a substance within the Society. The Canadian way of dealing with their more familiar 'two solitudes' might have some bearing on and even solutions for the problem of the two bibliographical solitudes.

for the use of others as well as themselves.This article, therefore, proposes a cooperative use of librarians' and bibliographers' skills, interests, and concerns in an attempt to develop a strategy for bridging the two bibliographical solitudes in the common cause of providing library catalogs and bibliographies to maximize the use of Canadian special collections of early books.
The Bureaucratic and the Idiosyncratic It might help to clarify the nature of the two solitudes if a somewhat oversimplified view of the librarian's attitude to books (as seen by a bibliographer) were juxtaposed against an equally oversimplified view of the bibliographer's attitude to books (as seen by a librarian).This writer can attest to the prevalence of such oversimplified views that emerge from papers written by Master of Library Science students upon their first looking into the literature of librarianship and of bibliography.They find (with a natural bias, of course, towards librarianship) that library science and its terminology are more standardized, conisistent, and well-organized for an explicit, though narrow range of purposes; and they find the work of bibliographers, as reflected in their societies' publications, idiosynrcratic, esoteric, and seemingly organized for a multitude of disparate and ill-assorted purposes.
These impressions are to a certain extent justified by the evidence provided in publications resulting from activities in the two fields.They are also explicable in terms of educational preparation of librarians and bibliographers as well as of the differing purposes they serve.Strongly regulated, widely dispersed, and highly institutionalized educational programs for preparing professional librarians for their work may be contrasted with peer-influenced, experience-driven, and weakly institutionalized pattems of educational preparation for the work of bibliographers.Departments of English, where textual scholarship is accepted as a (usually minor) part of the spectrum of literary and language studies, have attempted to institutionalize and formalize bibliographical studies at the graduate level, and this has left its mark on the 'main stream of tradition' as Fredson Bowers' calls it.However, the minor streams of tradition in bibliography have not yet been adequately treated by an allencompassing historian of bibliography.For librarians, Roy Stokes has made a valiant effort to characterize the field in historical terms in a ready reference book, 2 but his attempt to explicate the labels given to various types of bibliography seems to many a library science student to be more confusing than illuminating.Librarians tend to divide bibliography into kinds of bibliography -subject, author, national, etc. Bibliographers' inconsistent use of 'enumerative,' 'systematic,' 'critical,' 'analytical,' 'descriptive,' and even 'textual' (a literary critical term) or 'historical' (usually a librarian's term) for divisions of their field does not seem to be in harmony with the simpler divisions common in librarianship.For present purposes of characterization, it would not be too misleading to borrow the techniqlues of Matthew Amold and to focus our attention on the true mean of each of the two solitudes by labelling them with two extremes, those of bureaucracy and idiosyncrasy.
Librarians (or librarians-in-the-making) naturally view 'bibliography' as a product; bibliographers as a process.Librarians see the most important product as library catalogs, union catalogs, and sets of 'bibliographical data' for exchange in modern networks; they tend to view the idiosyncratic products of non-librarians as potential 'reference tools.' Bibliographers see the process as dictated by the need for non-bibliographical productsthe 'definitive text' of a literary work, an article or a book on some historical event or chain of events (and not only in the field of publishing or booktrade history), or an article or book for which careful evaluation is needed of the process by which primary sources were generated.
To library catalogers all books are eqlual, and the application of a uniform standard in their description of every book is of paramount importance; to bibliographers each book deserves the treatment necessary for the particular purpose for which their description is to be used, and these purposes differ tremendously.As a consequence, it seems certain that librarians know too little about the broader reaches of 'bibliography' other than their professional tailoring of it for cataloging and reference work, and bibliographers know too little about the broader reaches of librarianship other than the custodial and service elements that are immediately recognizable by the user of libraries.Bibliographers have not been as understanding of librarianship in their published papers as a librarian would like,3 and librarians (pace Stokes) have not shown as synoptic a view of the whole field of bjibliography as practising bibliographers would expect.
The most thoughtful and synoptic treatment of the interrelationship between the work of librarians and bibliographers is probably G. Thomas Tanselle's 'Descriptive Bibliography and Library Cataloging.'*He uses three sets of paired concepts (works vs. books, reference bibliography vs. physical bibliography, and enumeration vs. detail) to bring 'descriptive' bibliographers and 'descriptive' and 'subject' catalogers into mutual understanding of each other's preparation, use, and evaluation of catalogs and bibliographies.These three sets draw heavily on the theoretical traditions represented by Seymour Lubetzky, Lloyd Hibberd (or perhaps Rolf Du Rietz), and the Hand Printed Books (HPB) Project.`Tanselle's use of the concepts leads him to a statement and a footnote that provides us with an opportunity to seize on a common departure point for enlisting the common support of librarians and bibliographers : As the detail in a reference or physical bibliography becomes less, the entries in the two come to resemble each other more and more, and for that reason I have employed he same word 'enumerative' to refer to the lack of detail in either case.Similarly, both catalogs and bibliographies which lack detail can be called 'checklists' or 'handlists' but their differing functions remain as before.(Reprint, p. 48, and a part of footnote 19.) The strategy proposed here is to attempt to standardize the minimal entry for a checktlist (drawing upon a century or more of librarians' experience of attempting to standardize an optimal entry) and then to add to each entry extra information (according to an 'ingressive principle') that will draw upon the librarian's experience with 'reference bibliography' and the bibliographer's experience with 'physical bibliography.' The ingressive principle, first enunciated in I98I, 6 has not yet been definitively formulated, but this should not be surprising, as the 'degressive principle' (of which it is the obverse) was first enunciated in 1909 by Falconer Madan and has also never been unambiguously formulated.Fredson Bowers very successfully attempted to synthesize the principles of bibliographical DESCRIPTION as practised by the 'main stream of tradition,' and in the process of setting standards for 'an ideal intensive bibliography,' he alluded perfunctorily to the degressive principle thus: 'According to the material treated, and the purpose and means of a bibliography, omission or condensation of details can thereupon be made as seems necessary or advisable.'The HPB IrOject (from which the Western Hemisphere Short Title Catalog Project described below has been derived) attempted to synthesize the principles of bibliographical IDENTIFICATION (nOt description) from the Ang;lo-American tradition of short-title cataloging.In mimicry of Bowers, the ingressive principle was described in the words he used for the degressive principle but with 'addition or augmentation' substituted for 'omission or condensation.'Additionsor augmentations may be inserted before or after the short-title catalog (sTc) entry, and these annotations in practice seem to fall naturally into two categories: those before the entry provide details about the 'work,' those that follow provide details about the 'book' (cf. the distinction between reference and physical bibliography).
To illustrate how the ingressive principle may be applied in practice, an entry from a WHSTC bibliography of Lahontan is juxtaposed with the first of seven entries in Greenly's standard bibliography.7These seven collectively are equivalent to the one WHSTC entry.The WHSTC annotation about the work could be augmented with information about illustrations (as in Greenly), or an annotation after the entry could include notes on particular copies or bibliographical information similar to that in Greenly's 'Collation' notes.Those that do appear emphasize the dictionary contained in the publications, and the bibliographical information is confined to identifying the 'angel' edition and the 'ornament' and 'globe' issues of the later edition.The first entry in Greenly's bibliography of Lahontan.
In Vol.I of one of the JCB copies are found six leaves, different paper, con- taining "L'Autheur, au Lecteur . .." comprising a complaint and three pages of errata, dated at beginning of section, "LA Londres, ce 26.Novembre V. [ieux] Stile."The presence of this address in a copy of the Angel edition indicates priority of that edition over the Ornament edition.Inasmuch as the errata are for the first two volumes only, it is probable that the Supliment or Vol. 3 was issued after Vols.I and 2. See also Sabin 38635, note.[Line] M.DCCIII.
Signa~tures.A-I in twelves, K in ten.
(Reproduced by permission of the Bibliographical Society of America)

Bibliographical Control of Early Books
If we confine our attention to early books (defined roughly as the intellectual records surviving from the fifteenth to the eighteenth centuries), we shall find that the proportion of books and manus cripts of Canadian interest dating from this period is so small, in comparison with those of the nineteenth and twentieth centuries, that the feasibility of attempting their bibliographical control by tht•e proposed strategy is very high.In confining our attention to this period, however, we may discover yet a third bibliographical solitude -that of the antiqluarian bookdealer.Our proposed strategy should, however, also help to bridge the gaps between this solitude and those of the librarian and bibliographer.
In comparison with European and American libraries, Canadian libraries have very modest collections of bookts of the fifteenth to eighteenth centuries.Nevertheless, these collections, when properly analyzed, frequently contain surprising strengths and unusual opportunities for research, at least for preliminary research that might reqluire visits to other countries at a later stage of investigation.For potential researchers, however, those strengths and those opportunities are not easily discoverable because neither librarian nor bibliographer has identified, described, or analyzed the collections in such a way as to bring them to scholarly attention.
The librarian's general solution to the problem of making bookts in a special collection accessible is to incorporate each item into the general catalog of the library, and by reporting the descriptions tO CANUC (the Canadian Union Catalog of Books), make them known outside the individual library.The creation of a general union catalog solves the problem of identifying and finding copies of individual books (or 'known items,' as the jargon has it), but it does not solve the problem of identifying a special collection except, perhaps, those that the collocative devices of a union catalog actually emphasize, i.e., author collections.But there are grave limitations to an author listing within a union catalog.'Analytics' (to use the librarians' jargon) are seldom found in the originating library's catalog from which the union catalog is made, and for economy are in any case usually edited out in the union catalog.Similarly, subject cataloging is also diminished.Thus a particular author's work that appears in a multiple-author book or in another author's publication seldom appears in the listing.
Nevertheless, the strengths and weaknesses of an author collection are fairly easily ascertained by flipping over the cards or browsing the machinereadable files of a union catalog.The frequent recurrence of a particular library symbol suggests very quicktly the relative strength of one library collection over another.The best way of converting such an impression of relative strengths into a more definitive judgment is to compare the items in the file with a'standard' author bibliography, if one exists.In the absence of such a bibliography, the union catalog becomes the best substitute we can hope to find.Unfortunately, most author bibliographies created by non-librarians tend to be organized on completely different principles from those followed by library catalogers, and the models tend to differ considerably from one another, according to the different purpose or purposes each bibliography is intended to serve.

David Humîe
For a concrete example, let us take David Hume, surely a powerful enough influence on the thinking of our ancestors to justify the choice.Qualitative measures would in fact reinforce this conclusion.
The scholarly use to which this collection could be put is extremely limited, however, as the availability of modem editions of most of Hume's works suggests that major interest in early editions might be confined largely to gathering evidence for textual revision of modem texts.The bibliographical history of the early editions also has, however, some bearing on the study of the reception of Hume's ideas.Similar limitations also apply to most collections devoted solely to one man's monographic works.Unless intellectually related books by other authors are identified and perhaps described (or their relationships identified and described), the collection will not attract the scholarly attention it deserves.McGill has an interesting collection of what might be called 'Humeana' (i.e., bookts containing works relating to Hume), but apart from a small collection shelved close to the Hume items, it is scattered throughout the library and the library's general catalogs.Nevertheless, the very identification of a strong author collection should warrant the description of the collection in the form of a checktlist, and the creation of this checklist could prompt the compiling of a similar catalog of such 'Humeana' as could be found in McGill.Unfortunately, this latter ktind of catalog is more in line with the characteristic work of an idiosyncratic bibliographer rather than a bureaucratic librarian.If, by some circumstance, the complementary second catalog were created by either a librarian or a bibliographer, it might encourage the librarian to compare the collection with those in other libraries, thus producing a desiderata list for the acquisitions librarian and a research tool for the scholar.It is a fairly simple matter to compare an author checklist with Nuc, but a similar comparison of an 'ana' list requires rather tedious consultation of a large number of volumes of Nuc.
The very tediousness of manually searching Nuc for material widely scattered throughout its 746 volumes suggests that a facilitating mechanism is sorely needed to permit such comparisons to be made by a special collections librarian working within the bureaucracy rather than a volunteer bibliographer with particular interests in mind.In an effort to overcome the necessity for time-consuming and expensive manual labour involved in such daunting enterprises, a prototype editorial, retrieval, and display system has gradually been developed over the past five years.

The WHSTC Project
The WHSTC (Westem Hemisphere Short Title Catalog) Project based in the School of Library and Information Science, University of Westem Ontario, has as its primary aim the bibliographical control of books published before 1801 that (a) were printed in the Westem Hemisphere, (b) were printed elsewhere but contain material about the Westem Hemisphere, and (c) are held in Westem Hemisphere libraries.The machine-readable files of the WHSTC SYStem consist of 'main files,' containing short-title catalog ( sT c) records made from first-hand examination of the books; 'control files,' which equate these records with Nuc records and identify libraries holding copies of the books; 'qluery files,' which, through the control files, select and display records (together with annotations) in predetermiined combinations and order of presentation; and 'browsing files,' which help the user of the system retrieve individual records for incorporation into qluery files.Query files may be created from library catalogs, bibliographies, or collections of actual books, or a combination of all three.The files of the WHSTC systemn are accessible through a retrieval program called NOBLE, which is designed to produce bibliographies or library catalogs, annotated or unannotated, in a variety of arrangements and at different stages of development.At their most primitive, these compilations may be used as checklists or finding lists.At their most sophisticated, they may supplement or complement, index, rearrange, aggregate, simplify, sophisticate, or even superseded existing bibliographies or library catalogs.At an appropriate stage of development, when its general use may transcend the particular use for which the compilers created a particular bibliography or library catalog, a printout is made and kept on file for 'ondemand' photocopying.This form of publication was thought to be superior to conventional publishing, as the machine-readable output will be improved as a result of periodic revision of the files from which it is generated and as new data becomes available in the WHUC (Westem Hemisphere Union Catalog) data base and the related files of the WHSTC systemI.
During 1982 ten w HST CBibliographies and eleven WHST C Library Catalogs were published in order to demonstrate possibilities and usefulness in Canada of the WHSTC systeml.During 1983 only a small number will be published, mainly to demonstrate the usefulness of such publications outside Canada (but within the Western Hemisphere).For instance, a library catalog of the pre-180I books in the Openbare Leeszaal van Curaçao and a bibliography of Caribbeana, 1506-I727, in Dutch have been published to highlight neglected historical material that can be found in the Dutch Caribbean.A bibliography of French and Spanish books about Santo Domingo (both the Dominican Republic and Haiti) is also in preparation.
The WHSTc systemn was developed with the aid of a research grant (No. 4Io-78-oll2) from the Social Sciences and Humanities Research Council of Canada.This enabled Canadiana in French to be incorporated into the Western Hemisphere Union Catalog (WHUC) data base.It also enabled a browsing and retrieval system to be designed and implemented to form the basis for the system for producing bibliographies and library catalogs.The WHSTC system is intended to be capable of cooperative development, where users (bibliographers, librarians, scholars, students) may, by following some simple procedures, initiate a bibliography or library catalog of their own or modify one already in the system.Users of the bibliographies (IssN 07Iz-9289) or library catalogs (IssN 0717-929)7) in the two series of on-demand publications are therefore invited to comment and criticize, make suggestions for improvement, or even request that supplementary and co~mplementary listings be created for maximizing the usefulness of those already published.Those printouts that have reached publication status are listed in the preface in each on-demand publication, together with the original creation date of each list and the latest revision date.

WHSTC Bibliograpl2ies and Library Catalogs AUTHOR LISTS
The Hume catalog" lists the McGill holdings in roughly the same order as the filing order of Nuc.This enables items in Nuc (and/or Todd) that are not in McGill to be noted in the appropriate place in the printout of sT c entries.In augmenting the catalog in this way, items not in McGill can be identified by Nuc addresses or by Todd's reference number.The sTc entry is identified by means of a WHUc address of twelve characters, which incorporates the Nuc address of eight characters.For those forty-one items that are in Nuc, the sTc entry thus includes information that will lead the reader to further information contained in the regular library catalog card reproduced in Nuc.When an asterisk occurs in the ninth character position in the WHUC address, the reader may conclude that the item is not in Nuc in that particular form.Thirty-eight of the seventy-nine items in the Hume catalog are not entered separately in Nuc, although a few are subsumed in another, usually multivolume entry.
The library holdings given in the Hume entries in Nuc can be quickly listed and ranked according to numbers of items in a particular library.Of the 120 or so libraries listed as holding eighteenth-century editions, only 12 have Io or more books.Harvard (MH) has 41 of the nIo or so eighteenth-century bookts, Yale (CtY) 31, University of Illinois (IU) 21, and University of Texas (TxU) 20.Such figures are sufficient to demonstrate that McGill clearly has the largest collection of Hume in North America.Could the same be said for its Humeana?A wHSTC checklist is an obvious desideratum for finding out.
The figures demonstrate that a comprehensive bibliography to augment Todd's very fine, but deliberately circumscribed one could be initiated at McGill.It could probably be completed in Scotland (where the finest collection in the world is to be found), but collections elsewhere may also have to be consulted.This bibliography could be developed within the framework of Bowers' 'ideal intensive bibliography,' but the WHSTC systemn makes it possible to publish an interim comprehensive checklist, which can be augmented in stages according to the ingressive principle.When augmented to form a true bibliography (in Tanselle's terms), the wHSTc Bibliography would optimize the use of all Hume collections, not just the major one in North Amneri ca.
Hume has been used as a Canadian example of an author collection that could benefit from the compilation of an 'ana' library catalog, which might be used later as the basis for a bibliography of early books other than an 'author bibliography.'Other author collections in Canada could (and should) be identified for similar treatment.They need not be as comprehensive as the Hume collection at McG;ill, for it may very well be more important to identify special strengths within that author's canon or within a particular period of publication.John Milton can be taken as an example.
The G. William Stuart Collection of Milton and Miltoniana at the University of Western Ontario (CaOLU) is by far the best in Canada, but WHSTC Library Catalogs for much smaller collections at McMaster University (CaOHM), Queen's University (CaOKQ), and the School of Library and Information Science, University of Western Ontario (CaOLUI), have demonstrated that special characteristics make them of special interest to scholars.
The McMaster collection is the fourth largest collection of early editions of Milton in Canada, but is comparable with the Western (CaOLU) collection only where the CaOLU collection is weakest -during the last quarter of the eighteenth century (twenty-seven items as against twenty-five in CaOLU).A bibliography of collections of English poets and of English dramatists was appended to the WHSTC Library Catalog'o to provide a context for evaluating the research potential of the collection.(The major interest would be study of the publishing of literary texts after the House of Lords' copyright decision of 1774-) The Queen's collection is very small (even though it is the fifth largest in Canada), but it has a unique strength among its thirteen titles (or sixteen 'bibliographically distinct volumes,' to use the wHSTC unit of description).The earlier part of the collection forms a nucleus for a specialized ktind of Milton collection -the letters, especially the letters of state.To provide a context for this special strength, a checklist of editions up to 1738 of Milton's letters of state was added to the catalog of Queen's holdings."Analysis of the second and third largest collections in the University of Alberta (CaAEU) and the University of Toronto ( CaOTU) might also highlight a special strength if a WHSTC Library Catalog were compiled.
The School of Library and Information Science, University of Westem Ontario (CaOLUI), collection (now part of the Stuart Collection at CaOLU) was considerably larger than the Queen's collection, but analysis showed that it would be much more valuable as a research collection if it were added to the largest collection in Canada than if it were maintained separately.A WHSTC Library Catalogl2 CStablished some unique characteristics that would help to strengthen the larger collection.Some items not in CaOLU were obvious examples, but by far the most interesting (and rare) items were two copies of a French translation of Paradise Lost and Paradise Regain'd dated 1743.•They in fact proved to be two editions on two different batches of paper, one with watermarks dated I749, the other with watermarks dated 1756.Another edition in the Alexander Tumbull Library, Wellington, New Zealand, was subseqluently found to be on paper watermarked I746 and 1747.One wonders whether the two books noted in Nuc are different editions or not.Even ~three editions over a period of thirteen years will be of special interest to the historian who is yet to write the history of Milton in translation.
The remarkable strength of the Westem collection in the very earliest editions and the prior existence of a checklist"3 made it seem best to attempt to compile a WHSTC Bibliography rather than a Library Catalog.This publication'4 identifies eighty-eight items printed in Milton's lifetime, fifty-four of which are in Westem.An index of printers and publishers provides a springboard for certain kinds of study that only the original editions can facilitate.The bibliography can be used as a checklist for further establishing where supplementary material can be found.
From the information in existing checklists, it should be clear that Westem is liktely to hold the best collection of Miltoniana in Canada, so wHSTC Bibliographies could be solidly based upon their holdings, just as a checktlist of Humeana could be based on McGill holdings.'ANA/ AND SUBJECT LISTS The concept of a bibliography of 'ana' (that is, books relating to a person, place, etc.) goes well beyond a subject catalog or bibliography (books about a particular topic), as will be clear if a bibliography of Canada is compared with a bibliography of Canadiana.The fine library catalogs of Canadians in the Metropolitan Toronto Library (CaOTP) or of Laurentiana in the Bibliotheque nationale du Qu6bec (CaQMBN) that already exist"' or the private collector's catalog that serves as a catalog of the major part of the McGill University collection of Canadianal6 COuld be usefully summarized in WHSTC Library Catalogs, but it would be more useful to supplement them with a series of WHSTc Bibliographies.Two preliminary attempts at an author bibliography, for Louis Hennepin"7 and for the Baron de Lahontan,'s reveal serious weaknesses in Canadian library holdings.Thirty-one of the forty-four Hennepin entries are to be found in the Buffalo and Erie County Historical Society Library, but the largest collection in Canada (CaQMBN) has only fifteen items.Seventeen of the thirty-four Lahontan items are seemingly not available anywhere in Canada, and the largest collection (CaQMBN again) has only eight.However, no outstanding collection in the Uniited States has yet been identified.
Because of the weakness of Canadian collections, several draft checklists of Canadiana (arranged chronologically) still remain in the WHSTC files very far from the stage where on-demand publication is warranted.However, one easily separable series of books -the justly famous Jesuit Relations (16 32-I67 3) -has been assiduously collected by Canadian libraries.A WHST C Bibliography'" reduces to 5 7 entries the 132 in McCoy's very detailed bibliog-raphy2o S as T CSult of the standardizing practices of short-title cataloging.Over half of McCoy's entries are 'variations' that can be converted to succinct anno~tations and added to the sTc entry according to the ingressive principle.(Six of the WHSTC entries are also the result of variant states of particular formes that led McCoy to provide so many separate entries.)The conversion of McCoy's 'variations' into notes identifying editions, issues, and states in a WHSTC Bibliography provided an excellent example of how a detailed bibliography could be simplified without losing essential information (the degressive principle), and another wHSTC Bibliography, this time of eighteenthcentury Jesuit letters,21 prOvided an excellent example of how a sketchy bibli- ography22 COuld be amplified.The latter WHST C publication features the kind of annotation that will be of greatest use to those wanting to know the contents of each item; the former features the kind of annotation of greatest use to those wanting to know the textual and bibliographical significance of each item.An example of an entry from each bibliography has been supplied as an illustration.
The Jesuit Relations bibliography in fact provides a major challenge to both librarians and bibliographers to identify the uniqlue characteristics of holdings in Canadian libraries, as it is fairly clear that McCoy did not identify all variants.All except fifteen of the fifty-seven entries have been found in Can.adian libraries, but the precise issue or state of many of the items has not been  From WHSTC Bibliography No. 7, provisional index.
determined.The annotations in the WHSTC Bibliography will help the special collections librarian identify the particular issue or state of an individual leaf or forlme in a book in their collection -an example of a cooperative strategy that helps to bridge the two bibliographical solitudes.On the other hand, the major contribution of the other bibliography is indexing the contents of the eighteenth-century Lettres idifiantes.Canadiana, Americana, and 'ana' for others parts of the world are now readily identifiable, even though six volumes of the 1704-I776 series remain to be examined when copies are found.This 'reference bibliography' approach to the listing of the series poses a challenge to the bibliographer whose major interest might be in the textual value of different editions and different recensions of the text.Most modern writers refer to or cite later editions and recensions, mainly because of the rarity and arbitrary arrangement of the earlier ones.Bibliographers shlould be able to determine whether this has led to deterioration in the interpretation of the texts.

'PRINTERS AND PUBLISHERSl LISTS
The Jesuit Relations bibliography is, in some respects, merely a segment of a 'publisher' bibliography, for the Cramoisy name comes up in the imprint in all but nine of the fifty-seven entries.Similarly, the pattern of Le Clerc and Le Mercier names in the imprints of Lettre~s idifiantes could suggest that an investigation of the maj or publishers of Canadiana would be much facilitated by a checklist of their publications.This has not yet been attempted, but it would be most easily undertaken if the best collection in a Canadian library of a particular publisher or printer were first identified and described in a wHSTC Library Catalog.One example (only tenuously linked with Canadiana) of a checklist for the study of a particular printer or publisher might be cited.The School of Library and Information Science, University of Western Ontario (CaOLUI), has the most comprehensive collection in Canada of the two series called 'Bell's British Theatre.' Two hundred and seventy entries in sTc form and in conventional library catalog arrangement, with shelf-numbers appended, have been listed in a WHSTC Library Catalog.2 3 Many entries subsumed several copies or several variant editions, issues, or impressions.Three WHSTC Bibliographies have since been published24 tO provide the bibliographical context for the volumes in the collection.
The three Bell bibliographies revealed many new facts and suggested new interpretations of the motives and marketing strategies for Bell's theatrical publications.They also provided new evidence for his complex relationships with publishers of similar play collections.The various forms in which the plays were issued from 1780 to 1793 cast an interesting light on bookselling practices for popular material in the late eighteenth century.Of more interest to librarians is the discovery that many of the most useful copies of these later issues of the books are to be found in public libraries in unique combinations.The copies of two completely distinct aggregate issues in the Metropolitan Toronto Library (CaOTP) and the Hamilton Public Library (CaOH) are of particular value for elucidating Bell's practices.McMaster University (CaOHM) also has a distinctive copy of the extremely rare 1791-1793 'royal octavo' series.The checklists should, if used appropriately, also establish the special research value of many another collection.
Another example of an evolving series of WHSTC Library Catalogs of a printer or a publisher that will eventually lead to the creation of a WHSTC Bibliography can be cited.25 ThiS catalog of E1zevier imprints and Elzevieriana in CaOLUI was intended to be used by students studying significant contributors to the history of the book and fine printing.But the fact that the earliest examples of 'E1zeviers' were three items in their series of 'Republics' led to the identification of the largest collection in Canada (McGill) of these books on 'the climate, produce, religion, manners, civil and political govemment of these several states.126The WHSTC Library Catalog27 Of this collection draws attention to the likelihood that it once formed part of the collection of schoolbooks with which Peter Redpath was furnished while at school in Montreal in the 1830s.The Redpath collections in McGill contain many specialized collections that could be made more accessible to scholars by incorporation in a WHSTC Library Catalog.Meanwhile, a supplementary collection of Elzevier Republics at York University (CaOTY) is being studied, and eventually a WHSTC Bibliography (probably of Joannes de Laet, the author or editor of many of the books) may be published.De Laet's interest in producing the Republics also extended to accounts of the New World, including Canada.

'FORM'
OR /GENREI LISTS The Elzevier Republics could be classified as a subdivision by form of the output of the famous Elzevier family of printers and publishers.Form headings in the bureaucratic wo~rld of library catalogs are not very frequent and are always subservient to subject headings.Thus, any scholar requiring access in a general catalog to a collection of bookrs characterized by similarity in form (whether physical form, such as pamphlets, or intellectual fonrm, such as dictionaries or encyclopedias, or literary form, such as plays, poems, or sermons) finds it extremely difficult to identify such material without a 'standard bibliography.'Obviously, the WHSTC Projet should place high priority on the identification in Canadian libraries of collections strong in such material.A 'form' or 'genre' bibliography or library catalog may, of course, be a subset of an author bibliography too, and an example has already been published in each of the tWO WHSTC series.A catalog28 Of eighteenth-century editions in CaOLUI of the Spectator (periodical essays written by many authors, but conventionally attributed to the major contributors, Joseph Addison and Richard Steele) features annotations on the provenance of particular copies in that library.That the Spectator was considered to be highly suitable reading material for young ladies is but one deduction that one may draw from the index of previous owners.Examination of other copies in Canadian libraries may establish whether this is indeed true.
A bibliography of prose works by Aphra Behn29 draws attention to the literary forms to which that pioneering professional writer turned in the last five or six years of her interesting life.The extreme rarity of seventeenth or eighteenth-century editions in Canada and the astonishing geographical scatter of U.S. holdings makes the existence of a collection of microfilms in the University of Saskatchewan (CaSSU) and the identification of the strongest collection in the United States (at Yale University, CtY) of great value to scholars who are becoming increasingly interested in early women writers.
Such form or genre suabsets of an author bibliography can be complemented by form or genre bibliographies and library catalogs, in which the authors may be listed in an appended index.CaOLUI provides an examnple, in that it has the largest collection in Canada of early dictionaries, both subject dictionaries and lexical dictionaries.It is supplemented by a collection on the same campus in the D.B. Weldon Library (CaOLU).A WHSTC Library Catalog of general or encyclopedic dictionaries in CaOLUPo0 has been supplemented with one of the Encyclopédie mithodique (1782-1832) in CaOLU.n'The CaOLUI 'form' listing will be followed by• a listing of biographical dictionaries and books of collected biographies, and then by a listing of subject dictionaries classified by subject.The CaOLU listing will eventually be superseded by a wHSTc Bibliography in which information derived from other copies in other libraries and from external sources will be used to clarify the history of this complex work of about two hundred volumes.It will also be complemented with a catalog of CaOLU copies of the Encyclopédie of Diderot, D'Alembert, et al., upon which P~anckoucke based the immense work .
These subj ect dictionaries contain a large amount of interesting 'ana' of all kinds (including Canadiana) that has never been included in standard listings.They also contain epitomes of information about people, places, and so on, that could lead, from citation or allusion, to useful organization of WHSTC library catalogs and bibliographies, which will attract the attention of scholars and students to existing subject strengths in Canadian libraries.Librarians and bibliographers should be able to cooperate in providing such guides for the fruitful use of special collections in Canada.The subject headings of modern cataloging codes are not as helpful as contemporaneous labels.

Building on WHSTC Bibliograpl2ies and Library Catalogs
Enough examples should have been given above of kinds of checklists (with hints at possibilities for selectivity in augmentation or addition according to the 'ingressive principle') to suggest ways in which substantive bibliographies of early books could be developed.These bibliographies will be firmly based on an infrastructure of the specialized library catalogs of Canadian collections of early books.Checklists bare of annotation may be used to identify strengths in several libraries.32 Special uses to which such strong collections, whether large or small, might be put could be determiined by consultation between librarian and bibliographer.Decisions reached cooperatively would dictate just how much and what kind of augmentation or addition of descrip- tive detail wo~uld be appropriate.Interrelated bibliographies and library catalogs, drawing upon the common material in~ the w HUC data base, might be developed for different purposes and (idiosyncratic) use or for broader purposes and (bureaucratic) use.Bibliographies based upon major collections both inside and outside Canada would draw attention to the materials available in Canada and also guide the Canadian scholar to major resources elsewhere." In other words, a partnership between librarian and bibliographer could lead to the creation of access tools for special collections that would optimize the use of the country's rare books and special collections and at the same time solve some of the problems of the two (or three?) bibliographical solitudes.
)taionhas•c been• a~dumbrira~Ited aboiCve onT F. ~3., wh•en• we re•fe~r•red t;o McCor's 44•-4)8 (WHSÇTC 18 There are sets of the Angel edition consisting of Vols.I and 2 accompanied by the I703 edition of the Supliment; by the I704 Suite; and by the I704 I36. The best collection of eighteenth-century editions of Hume's writings in Canada is undoubtedly that in McGill University (CaQMM).Comparison of McGill's holdings with The National Union Catalog: Pre-1956 Imprints (hereafter N uc) and with William B. Todd's bibliography of first editions and closely related variationss establishes that 54% of the items in the two listings are in the McGill collection.If we ignore the popular History of England, 69% of the other writings are to be found in that collection.Even with this rough quantitative measure, it becomes obvious that McGill is a maj or research resource.