UC-NRLF 
 
 B 3 T2b il2 
 

 ■I .";' ',V.'/ •.i'A'-'ji'i 
 
 
 
 
 
 
 .■>i ..,->; 
 
 
 
 7., ...• 
 
 
^ 
 
 o 
 
 The Library Assistants' Association Series, 
 
 No. 3. 
 
 THE DEVEiiOPIVlENT Op 
 NOTATION IN CIiflSSIFIGflTION 
 
 BY 
 
 H. RUTHERFORD PURNELL, 
 
 Of the Croydon Public Libraries; 
 
 Honorary Editor of "The 
 
 Library Assistant." 
 
 CROYDON. 
 
 191 1. Price Threepence. 
 
t^°^ 
 
THE DEVELOPMENT OF NOTATION IN 
 CLASSIFICATION. 
 
 By H. RUTHERFORD PURNELL, 
 Of the Croydon Public Libraries ; 
 
 Hon. Editor of " The Library ?\ssistant." 
 
 I owe an apology to this meeting to-night for the 
 apparent dryness of the subject with which I propose to 
 deal. What could be more uninteresting than the subject of 
 notation, library pressmarks, classification numbers, or 
 whatever other title you may wish to give to the finding 
 numbers with which books are marked? Yet if studied in 
 relation to the classification of books and its historical 
 development an interest can be found even for these dry 
 bones of the librarian's duties. The ways of arranging 
 books are infinite in number, and the methods of showing 
 their arrangement vary from a simple consecutive numbering 
 from one onwards, to an elaborate sj-stem of marking that 
 may include the use of all the symbols used in the art of 
 writing. 
 
 In dealing with notation, as with classification, it is 
 necessary to begin very early. We learn that in the 
 libraries that existed before books began to take anything 
 approaching the form of writing on paper, when thej' were 
 
 •A Paper read before the Library Assistants' Association, at the North 
 Islington Public Library, on Wednesday, January 11, 1911. 
 
 224G65 
 
merely engravings on stone or scratchings on clay slabs, the 
 stones or bricks were arranged by subject in specially made 
 cases. It is not difficult to realise that some means *\'ere 
 taken to show the position of each, so that when they were 
 removed for consultation they could be returned to their 
 proper shelves. Though libraries flourished in the days of 
 ancient Egypt, Greece and Rome, they have left very 
 little traces of their existence. Their classifications, 
 consisting as they did merely of the arrangement of 
 books in broad classes such as Physics, Mathematics, 
 Metaphysics, Ethics and Politics, yet show the hoary antiquity 
 of the principle that books should be arranged by 
 subject rather than by any other method. This principle 
 has survived through many a dark age in the history of 
 books, and is seen when, towards the Middle Ages, an 
 interest again began to be taken in books, and the learning 
 that is to be gained from them. In these early days, at the 
 very beginning of modern history, however, books were so 
 scarce as to need very little arrangement. But even where 
 libraries consisted simply in reading desks to which the 
 books were chained, convenience would suggest that all 
 those dealing with theology should be together, and 
 similarly, those dealing with other subjects. It is not 
 possible to say for certain how the arrangement was 
 indicated, but probably the books were so few as to need no 
 such guide. It was when books began to multiply through the 
 invention of printing, that libraries of any size sprung up, 
 and with the accumulation of books, problems of their 
 arrangement from the first appeared. Classification as a 
 science is not confined to books ; though some librarians 
 may be apt to monopolize it to themselves as if they 
 invented it and perfected it, and to think that the 
 classification of other things, even of apples, potatoes, and 
 oranges in Mr. Brown's well-known example of the coster- 
 monger's barrow, does not matter. But classification is 
 necessary in almost every department of life ; and so, too, 
 notation is useful if not necessary wherever classification is 
 applied. For instance, in the science of chemistry, the 
 problems arising from the inquiry into the composition of 
 matter involve the naming of many elements and substances. 
 Where it is necessary to name a substance made up of 
 several constituents or show its combination with another 
 substance, the process would be very complex were a nota- 
 tion not designed to express each substance briefly. To give 
 a concrete example; water is composed of two parts of 
 hydrogen and one of oxygen. The chemist calls hydrogen H 
 
and oxygen O and the proportionate combination of the two 
 H2O. Similarly suphuric acid, consisting of two parts of 
 hydrogen, one of sulphur, and four of oxygen he calls H2SO4. 
 If he wants to add sonvo water to some of the latter he can say 
 H2O + H2SO+. The result will most certainly be in the 
 nature of an explosion, but he has expressed it simply. Nota- 
 tion is used in manj' other departments of life and takes many 
 different forms, as for instance in music, most complicated 
 sounds are denoted by a few black spots or circles with 
 straight lines running from them and curly ones in between 
 ■ — I am not a musician. 
 
 Now your clever modern librarian, in making a first 
 attempt to give a notation to the arrangement of his books 
 went to work in rather a curious way. He knew the main 
 subjects dealt with in books, and built a series of cases to 
 contain his books. Then he went to the cases and said, this 
 case must be for theology and we will call it A. This one 
 must contain history, let us name it B. The next one he 
 set aside for science and called it C, and so on for as manj'' 
 cases as he had. Each case was rather big, and had a lot 
 of shelves in it, so as he wanted to show exactly where each 
 book was placed he numbered the shelves one, two, three, 
 and so on to the end of the case. Then it was easy to call 
 the first book on each shelf one, and the others two. three, 
 and four to the end of the shelf. This was simple : easily to 
 be understood by anyone looking for a particular book. If 
 told by the librarian or the catalogue that the book he 
 wanted was numbered A. 1.4, all he had to do was to go to 
 case A and take down the fourth book on the first shelf. 
 I am of opinion that this marking of the presses to show 
 the position of books was the commonest if not the only form 
 of arrangement in use in the early libraries of modern 
 times ; and there may be a reason for these pressmarks to 
 be found in the habit of chaining books so that they should 
 not be stolen. Many of the monastic libraries were so 
 arranged, and the whole of the oldest part of the Bodleian 
 Library is pressmarked on a similar principle, as well as a 
 number of the College Libraries. 
 
 Here I should like to say that, though there may 
 be better fields for the study of classification than the 
 Bodleian Library, I have yet to hear of one, and I need 
 go no further in considering the subject of notation than 
 this one library, except for purposes of comparison and 
 application. At the beginnings of its history as Bodley's 
 Library, the collection was classified as I have already 
 indicated, except that more than one case was devoted to 
 
each class, and the name of the subject groupings was added 
 to the shelf-mark. There were four classes which corres- 
 ponded with the University Faculties: Theology; Medicine, 
 including botany, natural history, etc.; Jurisprudence, 
 which included history and everything relating to the 
 government of society ; and Arts, a quite miscellaneous 
 section. A certain portion of the building was set aside for 
 each class, the cases lettered Theol. ABC, Medicine 
 ABC, etc., and the shelves of each case numbered 1 
 onwards. The complete notation of the books in this part 
 of the library is exemplified in the pressmark A.1.1. Th., 
 denoting the first book on the first shelf of the first 
 theological case. In any library where this system of 
 numbering the presses is adopted, everything goes quite 
 happily if the cases can all be filled up and no more books 
 are expected. But if the library is growing there soon 
 comes a time when the cases get filled up, and trouble begins. 
 Either the books must be entirely re-arranged and re-num- 
 bered, with spaces left at intervals for growth, or a new 
 series of cases must be started with a similar scheme of 
 classification, and when they are full a third series, and so 
 on. The classification would thereby be rendered practically 
 useless. This is exactly what happened at the Bodleian. 
 The cases and in fact the whole building became filled; new 
 wings were built and a snnilar classification adopted. The 
 books in the new portions were distinguished by an addition 
 to the pressmark of the name of the new part of the library 
 as in the case of the wing built by John Selden. Here the 
 pressmarks are A.1.1. Med. Seld., etc. 
 
 To illustrate the difficulties attending this shelf numera- 
 tion in a growing library, I might mention my experience in one 
 of the Oxford College Libraries. This library was very ably 
 classified by Edward Edwards, that sadly neglected pioneer of 
 the library movement. Unfortunately Edwards adopted the 
 shelf marking system. He numbered all the cases from 1 
 up to two or three hundred ; gave letters to the shelves, and 
 separate numbers to individual books ; so that the press- 
 marks are simply 2 a 1, 4 b 6, or 200 e 25 as the case may be. 
 To-day the library has become full to overflowing, and an 
 amusing expedient to gain space is to push back a shelf-full 
 of books and arrange another row in front, altering the 
 pressmarks to double letters (AA, BB, etc.) for the back 
 row, and retaining single letters for the front row. The 
 cases are all, besides being solidly built with good old 
 English oak, made very deep. In many instances, advan- 
 tage has been taken of this depth not only to double row 
 
the shelves, but to treble and even quadruple them. Each 
 alteration means, of course, turning up the books in the 
 catalogue and adding letters to the pressmarks, or, in case 
 of the transference of books to other shelves, altering the 
 whole of the pressmark. Many an hour have I spent in 
 alterations of this kind. It is, to me, a matter for wonder 
 that so great a library as the British Museum still retains 
 this antiquated form of pressmarking. 
 
 It would seem that there came a time in the history of 
 the Bodleian when any attempt to classify the books seemed 
 hopeless. They were coming in so rapidly that time or 
 inclination was wanting to classify them on the old system 
 with its ever recurring need for alteration or rearrangement. 
 In the year 1824 the extraordinary plan was adopted of 
 taking all the 8vo. accessions for one year, arranging them in 
 alphabetical order, and numbering them from one onwards. 
 The numeration thus appears as 1824.1, 1824.2, shortened to 
 24.1, 24.2, etc. It thus happens that a theological treatise 
 jostles alongside of a novel, the novel next to a medical work, 
 and that beside a biography without minding it in the least. 
 After a while even the alphabetical arrangement seems to 
 have dropped and the books were simply numbered as they 
 came in. This form of numbering was kept up until the year 
 1850, and the collection is preserved as it stands, though 
 it has now been banished from its aristocratic position in 
 the picture gallery to a more humble one in the basement of 
 the old Ashmolean Museum. There is one thing to be 
 observed about this peculiar arrangement that is of interest 
 for our present purpose. Whereas before, all the book- 
 marking was' in relation to the cases in which the books 
 were stored, in these year-books, as they are named for 
 short, the numbering is independent of the cases altogether. 
 As long as you can find the books of the year 1824, 
 wherever they may be, you can find any particular book of 
 which you have the number. This is a distinct departure 
 from previous methods of marking, and maj^ have suggested 
 the later developments in book arrangement. 
 
 A most interesting experiment tried, it is thought, 
 between 1850 and 1860, was a form of lettering applied to 
 labels affixed to the books to denote subjects, of which a few 
 specimens will give the best idea : — 
 
 Ev = English verse ; B == Botany; Lp = Latin prose ; 
 P — Prayers; Dr = Drama; Chr = Chronicles; PE = 
 Political economy. 
 
 Stored away in one of the lower rooms of the Bodleian 
 is a little collection of books that is, I venture to think, of 
 
something like historic importance in the annais of classifi- 
 cation. It is difficult to find any definite particulars of the 
 collection, but I imagine that Bodley's Librarian of the time, 
 or one of his assistants, began to see the absurdity of the 
 arrangement of accessions by year, and wished again to 
 classify the books bj'' subject. But instead of apportioning 
 certain shelves to the classes as of old, he gave numbers to 
 the classes and carried the numbers into the books according 
 to their subject. It is interesting to find that the classes 
 follow the main idea of the original classification of the 
 Library, but shew an increase in number. The scheme is: 
 1 Theology: 20 Medicine ; 30 Arts and Trades ; 35 Law ; 
 50 Mathematics and Physics; 55 History; 70 Miscellaneous 
 Literature ; 85 Poetry ; 90 Classics ; 95 Philology. The 
 date of this collection cannot as yet be fixed with certainty, 
 but a clue can be found in the dates of the books themselves. 
 The first book in Class 1 is dated 1844 ; the first in Class 30 
 is 1847 ; in 35, 1861, and in 95, 1856. I think there is little 
 doubt that the collection was started about 1850, and it 
 is not difficult to surmise from what follows that the 
 arrangement proved a success, and led to the developments 
 that we shall come to directly. I wonder whether any 
 earlier examples of classification by subject, with a notation 
 altogether apart from the shelves, are to be found. The 
 Decimal Classification was invented by Mr. Melvil Dewey 
 in 1873-1876. However early Mr. Dewey's experiments 
 began, I do not think they could have begun so early as this. 
 And yet the principle is identical. The small collection in 
 the Hope-Montagu room of the Bodleian represents a 
 revolution in book-classification. Hitherto the standard of 
 numbering was according to the cases in which books were 
 placed, technically known as fixed location. The little 
 collection of which I am speaking is surely one of the earliest 
 if not the earliest arranged on the principle of relative 
 location. 
 
 This early classification scheme went further than the 
 provision of a notation for subjects. It assigned as com- 
 plete a book number to each individual book as anj' shelf 
 mark could do, and that by a simple and effective means. 
 Books of different sizes were separated and the letters a, b, 
 or c given to the respective sizes. Then in each class and 
 size the books were numbered one onwards. In section 35, 
 law for example, there would be three sequences, 35 a 1 ; 
 35 b 1 ; and 35 c 1. Thus early was established a complete 
 notation, by which books were individualized, without the 
 necessity for their being assigned to any particular 
 
bookcases. The collection could be moved about at will 
 without necessitating any changes of pressmark, and what is 
 equally important it was expansive. If any sections became 
 full, the rest could be moved on to allow further space 
 ad infinitum, or at any rate to the full extent of the library 
 building. Even now additions can be made to it when any 
 valuable books of an early date are purchased by the library. 
 
 This brings us to a consideration of the later develop- 
 ments of notation, not only in the Bodleian, but also in the 
 libraries of the world. The simple notation just described 
 contams all the essentials of a book number, namely, a 
 short sign for each class ; and then a number for each book 
 in that class. That notation can be elaborated to show 
 more than this we shall see a little further on. In 1850 the 
 " year books " ceased and gave place to a classification 
 scheme similar in principle to the little collection in the 
 Hope Montagu Room, like it based on the original main 
 classes of the Library, but with an entirely different nota- 
 tion. In the first table is given a reprint, slightly abbreviated 
 of the " plan of classification " presented with the " Report 
 of the Committee of the Curators of the Bodleian Library," 
 in February, 1878. I have given the old classification 
 scheme, which was in use from 1860 or some time before, to 
 1880, because Mr. E. W. B. Nicholson, on becoming Bodley's 
 Librarian in 1882, used it as the basis of an extended scheme 
 which, in respect to minuteness of division rivals any of the 
 existing schemes, and has placed the Bodleian on a plane of 
 usefulness as high as any of the great libraries of the world. 
 
 I wish that I could give here a table of the Bodleian 
 Classification at present in use, because it is deserving of 
 close attention, and is at present unknown beyond the limits 
 of the library. The scheme is probably unsuitable for any 
 other library, but if studied in connection with the history 
 of the library is of extreme interest, both from the point of 
 view of the closeness of division carried out as necessitated 
 by the actual books in hand, and because its main classes 
 resemble the original arrangement to a sufficient extent to 
 show a continuity of classification during nearly the whole 
 history of the library. The commencement of an entirely 
 new numeration would have involved the clashing of the old 
 and nev/ shelf-marks. To reclassify the volumes marked 
 under the old scheme, would have necessitated more than 
 300,000 alterations in the catalogue. 
 
 Whilst the Bodleian Library was progressing in the 
 science of classification along its own lines, library science 
 in general was certainly not standing still. It is curious to note 
 
Fixed Location. 
 
 Main classes, corres- 
 ponding witli the Uni- 
 versity Faculties. 
 
 Theology 
 
 Medicine [Includes 
 botany, natural his- 
 tory, etc.] 
 
 Jurisprudence [In- 
 cludes history and 
 everything relating 
 to government] 
 
 'Arts [Miscellaneous] 
 
 Sample of notation 
 A 4.6.Th. 
 
 " Year books." 
 
 1824-1850. 
 The books of each year 
 arranged in alpha- 
 betical order and 
 numbered 1 on- 
 wards. 
 
 E.g. 1824.1 or 24.1 
 1824.2 
 1824.3 etc. 
 
 Relative location. 
 
 About 1850. 
 
 1 Theol. 
 
 20 Med. 
 
 30 Arts and trades 
 
 35 Law 
 
 50 Math. & Physics 
 
 70 Misc. Literature 
 
 85 Poetry 
 
 90 Classics 
 
 95 Philology 
 
 Developments in the Classification of the 
 Bodleian Library from its commencement 
 
 to 1880. 
 
 223 General Hist. 
 226 Great and Britain 
 Colonies 
 
 231 Army and Navy 
 
 232 Political Econ. 
 
 233 American Hist. 
 237 French Hist. 
 240 German Hist. 
 243 Spanish & Portu- 
 guese Hist. 
 
 246 Misc. Foreign Hist. 
 MISCELLANEOUS 
 LITERATURE. 
 250-288 
 260 Education 
 
 264 Logic 
 
 265 Moralphiiosophy 
 & Metaphysics 
 
 268 Miscellaneous 
 
 [incl. domestic 
 economy, games, 
 sports, jests] 
 270 English prose 
 274 Romanic prose 
 278 Germanic prose 
 280 English verse 
 285 Romanic verse 
 288 Germanic verse 
 
 nil Bibliography 
 
 257 Palaeography 
 
 249 \ 
 
 „_„ I Novels and ro- 
 
 „ci I mances. 
 
 CLASSICS. 
 
 290-300 
 
 290 Greek prose 
 
 291 Commentaries 
 
 292 Greek verse 
 
 293 Commentaries 
 
 294 Latin prose 
 
 296 Commentaries 
 
 297 Latin verse 
 
 298 Commentaries 
 
 299 Misc prose(Since 
 the Classic .^ge) 
 
 300 Misc. verse 
 PHILOLOGY. 
 
 301-305 
 
 301 Comparative 
 
 302 English 
 
 303 Foreign 
 
 304 Greek 
 
 305 Latin 
 
 MEDICINE. 
 
 
 151-165 
 
 151 
 
 Medicines 
 
 160 
 
 Surgery 
 
 165 
 
 Anatomy 
 
 ARTS & TRADES. 
 
 
 170-176 
 
 170 
 
 Painting, Drawing 
 
 
 and Engraving 
 
 172 
 
 Sculpture 
 
 173 
 
 Architecture 
 
 174 
 
 Music 
 
 175 
 
 Misc. Art 
 
 176 
 
 Trades 
 
 MATHEMATICS 
 
 AND PHYSICS. 
 
 
 181-198 
 
 181 
 
 Arith. and Algebra 
 
 182 
 
 Calculus 
 
 183 
 
 Geometry 
 
 184 
 
 Astronomy 
 
 185 
 
 Optics 
 
 186 
 
 Mechanics 
 
 187 
 
 Misc. Math. 
 
 188 
 
 Geology 
 
 189 
 
 Zoology 
 
 191 
 
 Agriculture and 
 
 
 botany 
 
 193 
 
 Chemistry 
 
 196 
 
 Electricity 
 
 198 
 
 Misc. Physics 
 
 Classification 
 
 scheme in use from 
 
 I860?- 1880. 
 
 THEOLOGY. 
 
 100-147 
 
 100 Sermons 
 
 101 Commentaries 
 110 Eccl. history 
 130 Controversies 
 133 Missions 
 138 Liturgies 
 141 Miscellaneous 
 147 Sacred Poetry 
 
 LAW. 
 
 ENGLISH 
 TOPOGRAPHY. 
 
 ENGLISH 
 DRAMA. 
 
 ATLASES. 
 
 NUMISMATA. 
 
 HISTORY. 
 
 201-246 
 
 Geography 
 
 Vo3'ages 
 travels 
 
 Biography, Let- 
 ters, etc. 
 
 Genealogj' and 
 Heraldry 
 
 Chronology 
 
 Ancient History 
 and Mythology 
 
 201 
 203 
 
 210 
 
 218 
 
 220 
 221 
 
 and 
 
that what happened at the Bodleian found a counterpart 
 in libraries in general, though it may be from different 
 causes. I have already indicated that fixed location seems to 
 have been general in the early days of libraries. At the passing 
 of the Public Libraries Act there was no scheme of arrange- 
 ment to follow. The writings of Edwards were practically 
 the only publications existing, and it seems that he had not 
 got beyond the idea of making the shelves the standard of 
 numbering. The libraries that adopted the fixed method of 
 arrangement would soon be in trouble from the unequal 
 growth of different sections. It is my belief that the want 
 of a classification notation helped toward the universality^ of 
 the adoption, in English public libraries at any rate, of the 
 so-called classification by main classes. I do not pretend to 
 say it was the chief cause of this system. That should be 
 sought rather in the poor financial provision for libraries, 
 which did not allows enough money to pay men of sufficient 
 ability to organize and arrange the collections scientifically. 
 In the circumstances there is no cause for surprise that the 
 registering of books did not get much beyond a simple 
 numbering from one onwards as they were bought, a 
 numbering which was used in the shelf arrangement. I 
 suppose these numbers can be called a notation, but certainly 
 not a classification notation. Quite a startling innovation 
 would be the classification into about eight or nine main 
 classes already mentioned, that were referred to by letters 
 A, B, C, etc., and that had each its numbering from one 
 onwards as the books arrived. A 1, A 2, A3 is certainly 
 more of a notation, but it doesn't carry one very far, since 
 in the section science, for instance, you can get a book on 
 the blue bottle fiy next to one on electricity, and that next to 
 ■one on the daisy or the planet Mars. Mr. Brown, in his 
 manual of classification, has poured much scorn on this 
 labour-saving method of numbering books, still only too 
 frequently to be found in English libraries. 
 
 Then came the real study of the subject of book 
 arrangement, signalised in print first by the Dewey Decimal 
 Classification, with a notation purely numerical ; then by 
 Cutter's " Expansive " scheme, with a notation consisting 
 entirely of letters; later by Mr. Brown's "Adjustable" 
 Classification, with its combination of letter and number, 
 and last, but not least, bringing it right down to date, by 
 Mr. Brown's "Subject" Classification. You are probably 
 all familiar with the features of each, but in order further 
 to illustrate my point I give in a second table a comparison 
 
10 
 
 c 
 
 c 
 
 
 
 •— 
 
 '^ 
 
 JC 
 
 <0 
 
 
 >♦— 
 
 
 
 «4- 
 
 (/5 
 
 
 
 en 
 
 
 (0 
 
 c 
 
 u 
 
 
 
 MM 
 
 u 
 
 <0 
 
 0i 
 
 D.I/) 
 
 u 
 
 u 
 
 c 
 
 c 
 
 V. 
 
 
 
 a 
 
 JC 
 
 O 
 
 ^m/ 
 
 JC 
 
 ^ 
 
 O t(«:= 
 '^ O <« 
 
 v,^ ,1^ ,j_> 
 
 o _ <u 
 /) o 
 
 c c 
 
 en T3 
 
 o 
 'So 
 
 O 
 
 1—5 CS 
 
 C3'-5 
 73S 
 
 So2i o 
 
 C .5i "H 
 
 ■ua)«CQ!*=c>j5a 
 
 a cu CQ w M a w J J E CQ 
 <; m w 6 - -^i J § z 6 X 
 
 u 
 
 
 > 
 
 
 
 
 c« 
 
 
 7-. 
 
 >> 
 
 < 
 
 "S, _ 
 
 tt.7 
 
 O n 
 
 H en CO 
 
 
 OCUQi!riX;Oy3twZSDtu< 
 
 (0 
 
 L. 
 
 F 
 
 o 
 
 
 j:: 
 
 u 
 
 d-' 
 
 
 
 
 
 
 to 
 
 c 
 
 
 
 
 
 «0 
 
 
 
 to 
 
 
 £ 
 
 £ 
 
 x: 
 
 
 
 u 
 
 u 
 
 Jo 
 
 UJ 2 S 1^^ ° 2 - < 
 
 QW^0.2^3=2a, 
 
 OJ^JS O-S « 03.- 
 
 aanoia-zDix 
 
 ■oc 
 
 ectcj 
 
 
 - u 
 
 ■£ p 
 
 G C "^ 
 
 S 2 
 
 1. ai 
 
 cn « « 
 
 o > « 
 
 i2 §-'J w 
 
 • ■:;; c o. r- ^, - 
 o c 
 
 C r- r- W w- .:-( 
 
 « J .2 5 c '^ 2 
 
 - - 4) lU 
 
 
 a = 
 
 ~K?3«SoC;53i:KO 
 
 <M <N C^^ !N M (M « c; 1/5 IC IC >n IC X o o ^- 
 
 c^t^i^t^i>i>i>i>t^i>r>c^L^r^ Q0O5O5 
 
 l>^ Jm (Jm kX^ 
 
 ;-5 S--S WJ 
 
 u cn XI ^ ^ 
 
 OSO. 
 
 >> 
 
 >> N 
 
 OS K 
 >> 
 
 " cn .:: 
 
 f i- *J 71 
 
 O 5 ? =:i 
 
 <" S'iS o j; cs 
 =" ^ o .y 5 o 5 f'^ 
 
 fc tuO<- JCQ~ 
 
 ^ S t3 OCJ: ii S 
 
 >>>>>>>> 
 
 Z cn 
 
 ■J3x 
 
 oooooooooo 
 oooooooooo 
 
 z 
 
 
 'o 
 
 < >, 
 
 u -a 
 
 O) 
 
 5^ 
 
 5 £= 
 
 "cs 
 
 U o 
 
 •n cn 
 
 3 
 
 ohS<;z 
 
 S3 
 
 
 
 5.2 ^.^ g.^ g-2 
 
 i o^ o 
 
 i O IN 
 
 I o 6 
 o — 
 1 e^ (N 
 
 <30 05 OS CJ> 
 
 •I*' o> oa o 
 e<i IN CO CO 
 
 6 6 -^ 6 
 
 CO irj O 'i* 
 
 IM OO CO C<5 
 
 E y u 
 
 c -^ S , 
 
 C 3 ^ 
 
 < Dh r- . 
 
 O -^ ri 
 
 <J5 Oi CTl 
 
 l> I> t^ 
 
 I cn cn 
 
 I P J= I 
 
 0) =: ti5 
 
 u cn cs - 
 
 ' -g §■■? -s = 
 
 ; cn " t" 5 c X 
 
 ' S p Y *J cn |p 
 
 'S rt 3 O O.iG 
 
 , oomxci. 
 
 r^ i> t^ t> i> 
 
 -r IT) CD t^ (30 05 
 
 O'i Oi 35 0> 05 CJ> 
 
 r^ i^ r^ t> t> r^ 
 
 G 5 
 
 .o-o 
 '35 - 
 
 o ^ 
 ■'7 M 
 
 u "x; 
 
 U);2 D- t^ 
 
 *3 «_ 
 
 ^ go's 
 
 « >. 
 
 s^i Sa^ 
 
 ;CU « 
 
 ^ CM 
 
 GO CO 
 
 CO CO 
 
 o ^ 
 
 CO CD 
 C^ IN 
 
 -*• Tf< 
 
 C30 OO 
 CO CO 
 
 t; -5 ^ '^• 
 
 I"! — 
 ^-^ t^ ^ 
 
 CO w 
 
 ■^co ^ • 
 
 4-> ,^ "O 
 
 ^ CO ^ 
 
 ^ CO 
 
 CO CO CO 
 
 -1- TJ" Tf 
 
 GO 00 OO 
 
 CO CO CO 
 
11 
 
 of four of the principal classification schemes, followed b>^ 
 a section from each to show the methods of division. 
 
 As already illustrated, classification notation can be, and 
 has been, carried further than the numbering of the subjects 
 of books so that when the numbers are carried on to the 
 books and the latter are arranged b}- those numbers, they 
 will be in a methodical order. It is not reallj' complete until 
 it has got as far as providing a number for each individual 
 book, as was accomplished by the fixed classification notation. 
 The futher signs are technically termed classification 
 auxiliaries. In the Bodleian Library this problem was 
 early solved by giving letters to different book sizes and 
 adding a number for each book. This was done in the little 
 collection in the Hope-Montague room that I have described 
 already. 50 a 1 for example would be the first mathematical 
 book of a size, say, from seven to nine inches in height. In 
 the old classification scheme, adopted about 1850, letters 
 were still used, but seemingh' only for the purpose of 
 separating the class number from the book number; because 
 we find that when a section began to get too large, it was 
 ended and a new letter used, to start a fresh numeration 
 from one onwards until that in its turn became too big. In 
 some of the sections are utilized all the letters of the alphabet 
 in their turn, from a to z, and then the letters are doubled. 
 100, sermons, a very big section, begins at 100a 1, 100a 2 and 
 so on down to five or six hundred; then begins 100b 1, b 2 
 etc, ; which is followed by c and all the letters dow^n to 100 
 z 1 onwards. After that the alphabet is doubled lOOaa 1, 100 
 bb 1, and so on. In Mr. Nicholson's scheme now in use, 
 letters are again used to denote the sizes shown by the table : 
 
 a books over 20 inches in height 
 
 b 15-20 
 
 c 12-15 
 
 d 9-12 
 
 e 7- 9 
 
 f 5- 7 
 
 g below 5 inches 
 
 Then the books are numbered from 1 onwards as they are 
 received into each section ; and since the Bodleian is one of 
 the four libraries that receive a copy of every book published 
 under the Copyright Act, this numbering practically ensures 
 an arrangement in chronological order, one extremely useful 
 for bringing into prominence the latest books on various 
 subjects. A table of numbers has been devised by Mr. W. S. 
 Biscoe, for securing a similar chronological arrangement in 
 
12 
 
 connection with classification. He gives to each century, or, 
 since the year 1800, to each decade, a letter, and further 
 marks separate years with numbers. For example, H 
 represents the years 1810-19. The year 1817 is marked H17. 
 Perhaps the most common form of arrangement in each 
 class is alphabetically by authors' names. In many libraries 
 the books are so arranged, but it is not thought necessary to 
 give any number beyond the class number because each 
 book as it comes into the library, is entered in an accession 
 book having a consecutive numbering from one onwards. 
 For charging purposes this accession number is used, some- 
 times in conjunction with the class number. In connection 
 with his Expansive Classification, or rather, as an auxiliary 
 to it, Mr. Cutter devised an alphabeting table to provide a 
 notation for the arrangement in each section. Its main 
 features are the use of one, two, or three letters of the 
 authors' names followed by numbers which are lower or 
 higher according to whether the letters they represent are 
 earlier or later in the alphabet. One letter is used for 
 consonants, excepting S for which two are always used, and 
 Sc for which three are used. For names beginning with 
 vowels two letters are always employed. The following 
 example will show how they are used : 
 
 C77 Cook Ir8 Ireland Sca7 Scarbrook 
 
 C83 Cousin Ir84 Ironside Sco6 Scott 
 
 C84 Crabb Is89 Istar Scr5 Scrofton 
 
 Without using numbers of considerable length, however, 
 it is often difficult to specify particular books. Mr. Jast 
 devised a scheme, which was published in "The Library 
 World," the basis of which was the use of three letters of 
 each authors' name followed by a number and then one or 
 two initials of the title. By this means individual books can 
 be marked; but even then a long number is often required. 
 The example 
 
 822 
 SHA4 
 .H 
 
 written fractionally for conveniejice in entering on the tag, 
 will give an idea of the marking. 822 is the Dewey number 
 for English Drama; Sha4 is the author mark for Shakspere, 
 and H for a work on Hamlet. If several works on Hamlet 
 are in the library, additional letters or numbers are added to 
 the last symbol for each work. 
 
13 
 
 Without wishing to add to the large number of book 
 marks already existing, I might suggest a simplified method 
 of providing a sufficientlj' approximate alphabetical order 
 together with a complete book number. The idea is simply 
 to use the first letter of the author's name where Mr. 
 Nicholson uses a size letter, and then to number the books 
 as they are added. That is to say, in a section like 620 
 engineering, the books would be numbered 620al, 620a2 for 
 books by Abbot and Armstrong on this subject. If there are 
 likely to be many books by authors whose names begin with 
 A, it would be a simple matter to arrange for the use of the 
 first two letters of their names. By using fairly close 
 classification divisions for the library this would be found to 
 provide a useful charging number. The accessions book 
 would give place to a combined accessions books and shelf 
 register, as at the Bodleian. From practical experience I 
 consider such a notation would be eminently useful for 
 public libraries even of the largest size. A method of 
 denoting sizes in connection with most classifications, is to 
 prefix a '" lower-case " f to the class number of books of folio 
 size (e.g. f843) ; a "lower-case" q for quarto books (e.g. 
 q942) and to leave ordinary octavos without anj^ extra mark. 
 Further letters can be used if a larger number of size 
 divisions is required. 
 
 I have alread}'^ shown that notation can be made to 
 mark more than the class. It can go much further than 
 denoting particular books. The best illustration of the 
 perfection to which it can be carried is probably to be found 
 in the expansion of the Decimal Classification devised by the 
 International Institute of Bibliography at Brussels. This 
 Institute, while preserving Dewey's notation, with modifica- 
 tions such as the dropping of cyphers and the decimal point 
 as used by him, has introduced an elaborate sj'stem of 
 symbols, by which almost any conceivable subject, its special 
 points of view, and even its relation to other subjects, can be 
 signified by notation. In the Institute's extended scheme, 
 form is expressed by a cj'pher within brackets ; place by 
 numbers within brackets ; time by dates (or shortened date 
 numbers) within inverted commas ; language bj^ the Dewey 
 number following the = sign ; relation to another topic by 
 the Dewey number for that topic following a colon ; and 
 relation to another section of the same subject by its abbre- 
 viated number following a hyphen. It is most difficult in a 
 few words to explain the complicated notation, but perhaps 
 an example will be sufficient to denote its usefulness. 
 
14 
 
 "A treatise in French on Asiatic labour in the Transvaal 
 mines in 1900 " would be denoted by the notation: 
 
 331 .625 (682) : 622 "1900" =4 
 
 331 = Work and workers 
 
 .62 = Foreign labour 
 
 .625 = Asiatic labour 
 
 (682) = Transvaal 
 
 : 622 = Mining 
 
 ■ "1900" = The year 1900 
 
 = 4 = The French language 
 
 I do not pretend that the full scheme is necessarjor possible 
 in a public or, in fact, any library. But for the arrangement 
 of the great catalogue that is being built up by the Institute, 
 it has been found invaluable. The example given is an 
 extreme case to show as many of the symbols in use as 
 possible. It may be mentioned that if a technical library 
 Avere marking this subject, Mining would come first in the 
 notation thus : 
 
 622 (682) : 331 .625 "1900" 
 
 that is, Transvaal mining as affected by Asiatic labour in the 
 year 1900. 
 
 The Bodleian classification has also in many directions 
 been extended to denote minute sub-divisions, as can be seen 
 in the Bibliography section. 2590 is the number for English 
 libraries; a particular book in a single library in a particular 
 place is denoted by a fractional notation as in the case of 
 
 2590 e 
 
 Oxford 
 1.6 
 
 2590 = English libraries 
 
 e = the size (7-9 inches in height) 
 
 Oxford = the place 
 
 1 = the particular library (the Bodleian) 
 
 6 = the number of the book. 
 
 The survey of notation would perhaps be incomplete 
 without some reference to its application to special library 
 classifications. By the kindness of Mr. P. E. Lewin, I am 
 able to give an account of a classification that he has devised 
 for the library of the Royal Colonial Institute. In this 
 library a geographical arrangement is all important and 
 forms the basis of the scheme. The notation is numerical for 
 place division, alphabetical for main classes of subjects, and 
 the classes are further divided by means of numbers. The 
 geographical numbering will be indicated by the following 
 brief table : — 
 
15 
 
 1 
 
 General. 
 
 
 5 
 
 South Africa. 
 
 11 
 
 British Empire. 
 
 
 6 
 
 America. 
 
 12 
 
 British possessions 
 
 in Europe. 
 
 62 
 
 Canada. 
 
 2 
 
 Asia. 
 
 
 7 
 
 West Indies. 
 
 3 
 
 Ceylon. 
 
 
 8 
 
 Australia. 
 
 4 
 
 Africa. 
 
 
 9 
 
 South Pacific. 
 
 There are about 900 geographical divisions in use, which are 
 divided as in the example: — 
 2 Asia. 
 
 22 India. 
 
 292 N. W. Frontier Province. 
 
 2924 Kohat Division. 
 
 It is only possible to give one or two examples to illustrate 
 the alphabetical division of subjects: — 
 A Topography, Description. D Bibliography. 
 B Discovery. E Ethnology. 
 
 C History. F Manners and customs, etc. 
 
 The classes are divided by figures as shewn in : — 
 M Political Questions M9 Defence 
 
 Ml Theory of Empire M96 Finance 
 
 Mil Imperialism M962 Statistics 
 
 M2 Relations of G. Britain with Colonies M963 Crime 
 
 The complete notation in use is written thus: — 59M85 
 (a book on the trade of Cape Colony). There are 600 sub- 
 divisions at present in use, and these, as well as the 
 geographical numbers, are capable of further division at will. 
 Certain form divisions are shown by a letter preceding .the 
 geographical number, as in the case of L to denote official 
 publications. 
 
 Another special classification scheme is that of the 
 Board of Education Library, published by Mr. A. E. 
 Twenty man in an official white paper. In this- scheme 
 letters are given to the main divisions of education : — 
 A General works F Special method 
 
 B Fundamental sciences G Hygiene and physical 
 C Educational theories education 
 
 D Organization K Textbooks 
 
 E Buildings 
 
 The further division is illustrated in the following 
 examples : — 
 
 COO General C05.3 Theory of Instruction 
 
 COl Principles of Educ. C05.3 Theory of Infant 
 
 Instruction 
 COl.l Systematic C05.31 „ „ Instruction 
 
 pedagogy of Defectives 
 
 C02 Aims C05.321 Blind 
 
 COS Internal economy C09 Educational Fiction. 
 
16 
 
 There are no doubt many other special classifications, such 
 as, for example, that of the British Library of Political 
 Science, the tables of which haye been published; but it is 
 impossible even to name them all. 
 
 I do not pretend to have presented an exhaustive treatise 
 on notation, but rather to have suggested its usefulness and 
 perhaps its importance. Sometimes it is necessary to travel 
 far in search of information relating to all the phases of a 
 subject in library economy. For instance, in studying book- 
 charging methods one would probably have to visit many 
 libraries in order to see the various methods in use. But in 
 the case of classification and its notation one can go to the 
 Bodleian and see nearlj- every method that has been tried at 
 one time or another, and how each has given way to some 
 other form more suitable or more convenient to the time. 
 Further, at each new stage in marking the books, the various 
 collections have been retained, often in their original positions, 
 and always with the numbers unchanged, so that it is 
 possible to journey from room to room and see how the 
 library has developed with respect to book arrangement and 
 numeration, gradually progressing to one of the most minute 
 classifications in existence. It cannot be said that classifi- 
 cation has yet reached finality. There is still an opening for 
 the display of inventive genius in simplifying or extending 
 the usefulness of existing schemes, or even for the introduc- 
 tion of altogether new notations. When finality has been 
 reached, if that ever does happen, there is still room for 
 ingenuity in applying the multitude of symbols that can be 
 brought into use. 
 
 [In the discussion it was pointed out by Mr. C. J. Purnell, M.A., that the 
 various sheU'marks mentioned in the paper by no means exhausted those in 
 use in the Bodleian. Additions of booiis printed before 1640, other than the 
 incunabula which form a separate collection, are classed together with the 
 title of Antiquiora. Collections both of MSS. and printed books are still 
 called after their donors' names, and, in some cases, the names are retained 
 for press-marking the modern additions to the Library in the particular 
 subject in which the collectors specialized ; e g. Gough Additions for English 
 Topography ; Malone Additions for English drama. The former would be 
 written G. Adds., often shortened to G A., and would be followed by the 
 special topography class numbers. No one without actual experience in a 
 large libi'ary could realiate the difficulties in the way of a minute classification 
 of such a tide of books as flows into the British Museum, the Bodleian, and 
 such libraries. The continental theses alone involved an enormous amount 
 of difficult work. In reply to Mr J. D. Brown's remark that classification 
 notation was being overdone, he said that at their next meeting (at the 
 London Library) the Association would have an opportunity of seeing a 
 Library which has no notation other than the nan'.e of the subject. 
 
 A question was asked as to whether the fixed location method was 
 altogether doomed ; to which the reply was given that for a growing 
 library this method was useless.] 
 
FOURTEEN DAY USE 
 
 KETURN TO DESK FROM WHICH BORROWED 
 
 IBRAR' libra: 
 
 LD 21-100m-2,'55 
 (B139s22)476 
 
 General Library . 
 University of California 
 Berkeley 
 
( t 
 
 mmmmiunmmni 
 
 14 DAY USE 
 
 RETURN TO DESK FROM WHICH BORROWED 
 
 LIBRAEY oCHQOL LIERAR:^ ' 
 
 This book is due on the last date stamped below, or 
 
 on the date to which renewed. 
 
 Renewed books are subjea to immediate recall. 
 
 JUL 2 1968 
 
 
 APR 2 8 1969 
 
 
 .'IAN 9 a ]Q7C, 
 
 
 ■ ^ " — "' >.) — rO 1 U 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 , , General Library 
 LD 21-40m-10,'65 University of California 
 
 (F7763sl0)476 Berkeley 
 
 
 
. , ' ^.'f-^rs